It hums out from under the couch, working its way across the carpet.

To avoid other furniture, it builds a map of the living room.

The vacuum then works its way across the floor, and returns to its dock.

In order to do this, it has to know where the couch is.

It has to know where the toys on the ground are.

It has to know where the other furniture is.

It has to know if the dog is roaming in the same room.

Early models used in these devices navigated by bumping into things and turning around.

Today’s robot vacuums do something considerably more sophisticated.

They build and store a persistent map of the home, using LiDAR, cameras and microphones to record room dimensions, furniture placement, doorways and stairs.

Many also learn patterns of use over time, including which rooms get cleaned when and, by extension, when the house tends to be empty.

The map is what makes the features people actually bought the thing for possible, which is cleaning rooms on command.

But the map does not stay inside the vacuum.

It syncs to the manufacturer’s servers so the app can display it, and many of the best-selling robot vacuums in the U.S. are made by Chinese companies.

Even iRobot, the American brand that created the category, manufactures largely overseas.

For years, none of this drew much attention.

That changed in February, when Sammy Azdoufal, a Spanish software engineer, accidentally gained access to roughly 7,000 DJI Romo vacuums across 24 countries while building a custom controller for his own unit.

He was able to reach live camera feeds, microphone audio and stored floor plans.

DJI patched the flaw within days.

Funny enough, DJI is the same company the Federal Communications Commission (FCC) added to its Covered List for drones two months earlier.

There is no public evidence that any government has ever used this data, but Washington’s position is now that it does not have to wait for evidence.

On July 27, the White House issued a National Security Determination stating that all foreign-produced advanced robotic devices pose an "unacceptable risk" to US national security.

The next day, on July 28, the U.S. government banned the import, marketing and sale of new foreign-made mobile robots in America.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

Why Pass this Now?

The decision fits into a broader push to strengthen domestic manufacturing, expand reshoring capabilities, and reduce reliance on foreign technology in critical industries.

The Trump Administration’s AI Action Plan identifies robotics as a next-generation technology essential to our defense and national security.

This National Security Determination states:

Robotics technology is rapidly advancing, driven by enhancements in AI, sensing, and actuation technologies.

Advanced robotic devices will be critical to creating efficiencies in our economy, dominating on the battlefield, and securing our homeland.

To ensure the United States is able to take advantage of this generational technology, the United States must have an independent and resilient robotics supply chain and industrial base.

Robotics is now being treated similarly to semiconductors and drones.

And this timing also reflects where the industry stands today and how the government views the growth of robotics.

According to Omdia, roughly 13,300 humanoid robots shipped globally in 2025, with Chinese manufacturers AGIBOT and Unitree accounting for about 70% of those shipments.

Beyond just finished robots, much of the supply chain including actuators, reducers, sensors and rare-earth magnets also remains heavily concentrated in China.

Now the push is prioritizing American-made hardware and software in robotics.

Read the full National Security Determination below 👇

China Robotics National Security Ban

174KB ∙ PDF file

Download

Download

What the Ban Actually Covers

The ban focuses on mobile robots, not robotics as a whole.

Covered products under this ban include autonomous ground robots weighing more than 4.4 pounds that can sense their surroundings, connect to networks and navigate independently.

That includes new sales and marketing of humanoid robots, quadrupeds¹, warehouse robots (AGVs and AMRs)^2,3, and even robot vacuums such as the Roomba!

The rule does not apply to fixed industrial robotic arms, most collaborative robots unless mounted on a mobile platform, drones, automobiles or medical robots.

Now, the actual definition of “foreign-produced” depends on whether a product qualifies as a domestic end product under federal procurement rules, not simply where final assembly occurs.

That means a robot assembled in Texas could still be considered foreign-produced if most of its critical components originate overseas.

How aggressively regulators enforce this standard remains uncertain, but it creates a clear incentive for manufacturers to source more components domestically.

One important clarification is that this is not a China-specific rule.

The standard is whether a product qualifies as a domestic end product under federal rules, not whether it was made by a Chinese company.

That means products manufactured in countries like Canada or Mexico are also considered foreign-produced, even though those countries are close U.S. trade and defense partners.

As a result, simply moving final assembly to Mexico does not avoid the restriction, which is the precedent workaround that companies have pursued historically.

It’s equally important to understand what the rule doesn’t cover.

While the determination identifies components such as actuators, batteries, sensors, and end effectors as supply chain and national security concerns, the restriction applies only to finished mobile robots.

Individual components are not prohibited.

This differs from the past when the U.S. restricted foreign-made drones in 2025 — it explicitly included key components like motors, batteries, sensors and flight controllers.

But in this case, regulators highlighted components as a concern but stopped short of restricting them — the foreign picks-and-shovels players are still in play.

Lastly, you will notice in the determination that previously approved products remain authorized, and companies can apply for conditional approvals while they work toward moving manufacturing into the United States.

This way companies don’t have to trash current inventory and can work towards adjusting their supply chains and switching to new vendors.

Notes:

¹Quadrupeds = That 4-legged Robot Dog thing

²AGV = Automated Guided Vehicle

³AMR = Autonomous Mobile Robot

How Long Will the Ban Last?

The Federal Communications Commission’s (FCC) Covered List has tightened steadily since Congress created it through the Secure Networks Act of 2020.

Each update has broadened restrictions on technologies viewed as national security risks.

• March 2021: Huawei, ZTE, Hikvision, Dahua, and Hytera were added, targeting telecommunications, surveillance and communications equipment.

• November 2022: The FCC stopped authorizing new equipment from those companies, effectively preventing new products from being imported or sold in the U.S.

• 2021–Present: Congress appropriated roughly $5B to replace previously installed Huawei and ZTE equipment, effectively reimbursing small carriers for physically removing that gear.

• December 2025: Restrictions expanded to foreign-made drones and several critical drone components.

• July 2026: Mobile robots and foreign-made power inverters were added to the Covered List.

The broader takeaway is that the Covered List has consistently expanded over time, with no companies or product categories removed.

Reshoring American infrastructure is the push across all innovative physical technologies with potential security risks.

Companies to Benefit

Join our community to continue reading and researching the best robotics plays 👇

Read more

1. Executive summary & where I stand

2. Q2 earnings, August 4

3. Hardware stack & roadmap

4. ROCm

5. Partnerships

6. Financial projections

7. Valuation & sensitivities

8. Risks

9. Concluding thoughts

Subscribe now

Executive Summary & Where I Stand

This piece updates the AMD deep dive I published in November 2025 and the April valuation note. It does not re-explain the business from the ground up like the November piece that covers the product families and the earlier partnerships.

This piece addresses changes since, including the new partnership agreements, the Advancing AI 2026 event held last week, the revaluation of the CPU market for agentic compute and a re-forecast.

Disclosure: AMD is my largest holding, accumulated primarily through LEAPS from April through July 2025 with subsequent adds. Readers should weigh the analysis accordingly.

AMD trades near $520, a market cap of roughly $868B, up 144% YTD and 229% over the TTM period, and about 9% below the ATH of $584.73 on June 30.

My November Deep Dive estimated around $350 by the end of 2026 and $1,000+ by the end of 2029 — the shares passed $350 within six months. My revised base case implies $1,666 by 2030 at the midpoint, equivalent to a 29%+ CAGR over four and a half years.

For readers new to the name, AMD designs and sells:

• Core Processors (CPUs): For standard laptops, desktop computers, and giant corporate servers. Recently, the market has recognized a new value proposition for these chips — agentic AI.

• Graphics and AI Chips (GPUs & Accelerators): Hardware designed specifically for powering heavy AI workloads and rendering video games.

• Networking and Specialized Tech: Custom chips used for routing data and powering specialized electronics.

• Complete AI Platforms: Instead of just selling the individual chips and pieces, AMD has now assembled their processors, AI chips, and networking gear together into a giant, ready-to-use AI computing rack (Helios). Currently, they are the only major vendor building these complete systems entirely out of their own proprietary hardware.

Three developments shape the next eighteen months based on their product roadmap.

1.) Helios and the MI400 ramp:

AMD declared Helios in full production at the event, with first shipments in September 2026, revenue recognition beginning at the end of Q3, and volume building through Q4 into 1H 2027. 8-named customers, OpenAI, Meta and Anthropic among them, have committed to ~15GW of compute on the platform. The MI455X GPU accelerator and other chips in the MI400 product family are purpose-built to power the 72-GPU Helios rack, so this product rollout is somewhat synonymous regarding timing.

AMD needs to convert this commitment into shipped hardware over the next four quarters in order for revenue to be successfully reflected from Q3 onward.

2.) The server CPU cycle:

Agentic AI runs on CPUs due to the shift of workloads from a single prompt-and-response model to a dynamic decision-making loop that requires heavy orchestration, tool execution, and memory management.

AI servers historically carried roughly 1 CPU for every 8 GPUs, but that ratio has moved to about 1:4, and Intel’s CFO has said it “could converge to 1:1 or even tilt further in favor of CPUs.”

TrendForce models agentic deployments at 1:1 or denser, and a more speculative version of the argument has the ratio inverting to two or four CPUs per GPU as agent fleets scale.

The demand is already visible in Intel’s results. Intel’s revenue was up 25% YoY, its fastest growth since 2011, with data center up 59%, server prices up 27% and ten long-term supply agreements signed under supply constraint.

AMD holds a record 46%+ of server CPU revenue with a stated ambition of reaching more than 50% of the market. The Company has raised its own estimate of the 2030 server CPU market three times since November, and has guided server CPU revenue up 70%+ YoY for the current quarter.

Said differently, AMD’s estimate of the 2030 server CPU market has moved from around $60B in November, to $120B+ in May, to approximately $220B last week, alongside a more than 50% revenue share target against a reported 46.2% today.

Regarding timing of this general compute ramp, from what I’ve seen with Nvidia and other infrastructure compute stocks, things generally get priced-in three to nine months ahead of reported results, which suggests a portion of AMD’s 2027 P&L may be reflected in the shares during 2H2026 as Q3 and Q4 confirm the ramp of MI400 and Helios.

—> The illustrative share prices in sections 6 and 7 (financial projections, valuation & sensitivities) should be read with that lag in mind.

I expect AMD to deliver on this massive product rollout, but I do not assume it will be flawless. Helios is the most complex rack-scale system created to date, and AMD is not as experienced as Nvidia rolling something out at this scale with the anticipated demand.

The funny thing is that the performance of AMD could largely be dependent NOT on the company, but rather their suppliers and vendors responsible for integrating the hardware into the datacenter and other applications.

I still remain bullish on this rollout over the next 12 months.

Share

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

Q2 earnings, August 4

AMD reports Q2 on Tuesday, August 4 after market close.

Management provided no new financial guidance at the Advancing AI Summit.

Wall Street is expecting $11.3B in revenue and non-GAAP EPS of $1.61, slightly above the company's guidance of $11.2B in revenue and a 56% non-GAAP GM.

Consensus estimates imply +47% YoY of revenue growth.

My estimate based on my model later in this writeup is ~$11.5B of revenue and $1.65-1.70 of EPS. Server CPUs guide +70% YoY which accounts for most of the upside in my estimate, with continued MI355X demand contribution.

This will not re-rate the share price on its own… guidance will.

I expect guidance to be strong, since the first Helios revenue lands at the end of Q3 and is therefore partly inside the Q3 guide. I would not be surprised, though, if this gets pushed out into Q4 by the Company, as Lisa Su is always conservative.

If this gets pushed out into Q4, I don’t expect this upcoming earnings to be “shock and awe”, and I would expect AMD to selloff 10%+ after earnings. This will all depend on the commentary and Q&A AMD releases in their report and earnings call.

My view is that the November report is likely the larger event.

The Q3 print carries the first Helios revenue in reported results, and the Q4 guide is where the step-up in ramp becomes visible. August should produce a good quarter + guide, and November should produce a similar quality quarter, but could produce a monstrous guide that drives substantial model revisions in the market.

If management does end up guiding beyond a single quarter as I mentioned, while potentially framing 2027, or quantifies annual data center AI revenue, the reaction could be huge this quarter.

Share

Hardware Stack & Roadmap

MI455X compared with MI355X:

The new flagship is a full generational step, built on TSMC’s most advanced process, and carries 432GB of memory per GPU against 288GB on MI355X, an increase of 50%.

AMD also measures 34x higher token throughput and up to 18x lower cost per token than MI355X.

Higher token throughput and cost per token are specs directly tied to memory, which is the largest bottleneck in compute efficiency and scaling today.

During AI inference, models are often memory-bound rather than compute-bound, meaning the hardware spends more time moving data in and out of memory than performing calculations. As a result, memory bandwidth and capacity play a major role in determining token generation speed and cost. Memory capacity determines how much of a model fits on each chip and therefore how many chips a customer must buy.

More broadly, AI budgets are rising, cost per token is a primary purchasing criterion, and a generational improvement of that magnitude allows a customer to serve substantially more users on the same spend.

Lisa Su actually framed it in those terms at the Advancing AI Summit during her Keynote, citing approximately 18x more tokens per dollar.

Two additional products complete the MI400 family — the MI430X, configured for scientific and government computing, which anchors the DoE’s two new systems, and MI350P, an air-cooled card for conventional enterprise data centers capable of running very large models on a single unit.

Let’s compare the MI455X against the competition. Against Nvidia, on memory, MI455X holds 432GB per GPU against 288GB on Nvidia’s Vera Rubin, which lines up with the 50% advantage AMD announced. This is the specification that determines how much of a model fits per chip.

On manufacturing, MI455X is built on TSMC’s 2nm process while Rubin uses 3nm, which makes this the first generation in which AMD holds the more advanced node rather than trailing Nvidia!

AMD’s broader rack-level claims are more total compute, 50% more networking bandwidth out of the rack, and up to 30% more tokens per dollar, though Lisa Su’s own characterization was more conservative on stage when she mentioned “an average of 10% to 15% more performance than the competition”.

Despite this, Nvidia’s rack can reportedly pool memory between its CPUs and GPUs in a way AMD cannot currently match, which is important for inference workloads.

Nvidia is also improving on their predecessor technology — CoreWeave measured Vera Rubin delivering 10x the tokens per megawatt of Blackwell, so both companies are measuring large improvements against their prior hardware.

So to be clear, my investment case does not require AMD to outperform Nvidia across every workload. I don’t think that’s possible.

Helios:

This groundbreaking, highly complex rack is comprised of 72 GPUs, 18 server CPUs, AMD’s own networking cards and security processors, and liquid-cooling infrastructure, in a double-wide cabinet weighing roughly as much as two pickup trucks and drawing as much power as a small neighborhood, at an estimated $5–5.5M per rack.

Per rack, it carries more total memory and more memory bandwidth than Nvidia’s equivalent, which is the design choice underpinning AMD’s inference positioning.

The rack runs on a first-generation networking standard with hundreds of supporting components per rack, which is why I do not assume a flawless rollout.

Helios is a reference design rather than a machine AMD builds itself. HPE (Hewlett Packard) leads integration, with Supermicro, Lenovo and others assembling the racks, and the vendors installing them in customer data centers control the final steps.

So the rollout depends as much on that supplier and integration chain as it does on AMD, which is the dependency I flagged in the summary.

Lisa Su, following the keynote, stated that “declaring production means we are ready to ship.”

The Helios rollout could be bumpy.

The CPU:

Venice, the sixth EPYC generation, is the first server CPU on TSMC’s newest process, offering up to 256 cores per chip and up to 1.8x the performance of the outgoing Turin generation, which AMD characterizes as among the largest generational gains in EPYC’s history.

It ships in four configurations, including one designed as the host processor in AI racks and another aimed at what AMD described on stage as the “agentic data center,” a build-out with dedicated CPU tiers for running fleets of AI agents.

Turin continues to carry current financial results (previous EPYC generation), and is the basis for the +70% YoY server CPU guide, the reported 46.2% revenue share, and EPYC deployment across more than 60% of the Fortune 100.

As mentioned earlier in the writeup, the Company raised its own estimate of the 2030 server CPU market three times since November, and has moved from around $60B in November, to $120B+ in May, to approximately $220B last week, alongside a more than 50% revenue share target.

Not only is the market expanding, but CPUs are starting to look like they may be a key bottleneck in advancing agentic compute. This is a recipe for pricing power, and we have already begun to see it.

Intel’s Q2 results highlighted this shift, with client processor ASPs increasing 27% YoY despite unit volumes falling 8%. Their data center revenue also grew 59% YoY to $6.3B, and the company raised prices on several high-end Xeon processors while extending lead times from roughly two weeks to six months.

Management attributed the increases to “demand-based pricing actions.”

Both Intel and AMD have effectively sold out 2026 server CPU supply, with server CPU pricing up 10-20% since March.

AMD’s pricing story is slightly different I think. Rather than relying primarily on supply shortages, it’s gaining market share while selling a larger mix of premium EPYC processors.

AMD now generates 46.2% of x86 server CPU revenue on just 27% of industry units, reflecting its concentration in higher-value products. In my model, which we’ll go through in detail later, continued CPU pricing and product mix expansion are larger contributors to margin growth than my GPU assumptions.

Ryzen AI laptops:

AMD also expanded its AI PC portfolio with new Ryzen AI processors capable of supporting up to 192GB of memory, allowing larger AI models to run locally on a device rather than in the cloud.

Local AI reduces latency, improves data privacy and eliminates cloud inference costs, aligning with growing enterprise demand for on-device AI.

The launch also comes as businesses refresh PCs ahead of Windows 10 end-of-support, supporting AMD's push into the higher-margin premium laptop market.

Radeon:

AMD also introduced a new Radeon gaming GPU, continuing the recovery in its gaming business.

Gaming revenue reached $3.9B in FY2025, +51% YoY, driven by a new GPU product cycle and a multi-year agreement with Microsoft to co-develop custom chips for future Xbox consoles and handhelds.

While gaming is unlikely to be AMD's primary growth driver, the business provides a stable, contracted revenue stream that complements the faster-growing data center segment.

FPGAs:

The Xilinx FPGA portfolio is an area where I think AMD will have substantial growth down the road, specifically in robotics. I’ve been talking about this for over a year since their acquisition of Xilinx, the undisputed leader in FPGAs.

During the Advancing AI Summit, AMD introduced its Kria AI robotics platform, expanding the company's presence in physical AI. The platform combines CPU, GPU, NPU, and FPGA compute into a single development system designed for autonomous robotics and industrial AI applications.

AMD believes physical AI represents a significant long-term opportunity, and I hope they continue to have this vision, and start making it a priority going forward into future years as robotics and humanoid manufacturing accelerates.

Roadmap:

AMD has committed to an annual cadence of innovation, with the MI500 in 2027 described as its largest generational step to date, followed by MI600 in 2028, each accompanied by a new Helios generation and new EPYC series.

ROCm:

My November piece identified software as AMD’s principal weakness, which I still think is the case despite recently announced progress at the conference.

Nvidia’s CUDA has benefited from more than 15 years of ecosystem development and will continue to be the industry leading software through 2030, at least.

The primary update was the launch of ROCm.ai.

Rather than asking developers to learn AMD’s software directly, AMD is enabling the AI coding assistants developers already use (Claude, Cursor, Codex) to generate AMD-optimized code on their behalf.

So a developer will essentially describe the workload and the performance target, and the assistant handles the AMD-specific work.

Vamsi Boppana, AMD’s SVP of AI, described the goal as turning AI assistants into “ROCm superusers,” tools that can help accelerate software development and optimize AMD’s ecosystem.

AMD highlighted early examples of AI-assisted optimization, including one engineer using the system to tune 14,000 models, a task Boppana said would not have been feasible previously. In a live demonstration, AMD also showed AI-generated low-level code outperforming the existing human-written version by 38%.

AMD’s latest software release delivered a 3.3x improvement in inference and 2.4x improvement in training performance on identical hardware versus the prior release, driven entirely by software improvements.

The company also noted that its stack now integrates with major AI frameworks without custom configuration, with MI455X support available at launch.

Independent benchmarks place AMD at roughly 90-95% of Nvidia’s inference performance at 15-30% lower cost, while training remains 20–30% behind.

This gap in performance is less concerning given AMD’s view that inference will represent roughly 60% of AI compute demand by 2026 and is the faster-growing workload, where memory capacity and cost per token are key purchasing factors.

One notable reference to AMD’s software stack capabilities during the conference was when Anthropic co-founder and Chief Compute Officer, Tom Brown, shared that a single engineer hooked up Claude to an AMD MI355X rack, tasked the AI model with bringing up and optimizing the hardware, and left it running over the weekend.

By Monday morning (without human intervention) they were met with a real performance graph of Anthropic's leading model steadily climbing up and up throughout the weekend.

Because these automated enhancements flow directly into the open-source ecosystem, AMD customers can benefit from those improvements broadly. AMD has made meaningful software progress over the past eighteen months, although Nvidia’s ecosystem remains more mature.

I am hopeful AMD can continue closing this gap with CUDA.

SemiAnalysis argues that AMD’s infrastructure for external developers to test and optimize workloads on its newest hardware remains underdeveloped, with improvements expected in October. An important validation point will be independent benchmarks confirming whether MI455X can deliver near-Nvidia inference performance at a lower cost per token.

Partnerships

OpenAI — 6 GW committed, $78.8B forecast below (October 2025)

OpenAI’s 6 GW agreement represents the largest commitment to date.

The first GW is expected to begin deploying in 2H 2026, with Helios coming online in Q4 2026. OpenAI also holds a warrant for up to 160M AMD shares at $0.01, tied to GW deployment milestones and a $600 share price.

AMD has described the agreement as worth “tens of billions.” In the forecast, I included 5.25 of the 6 GW within the forecast period, with the remaining capacity extending beyond 2030.

Meta — up to 6 GW, $75.0B forecast below (February 2026)

Meta’s agreement includes a custom accelerator variant, Venice CPUs, first shipments beginning in 2H 2026 and a similar warrant structure.

Meta’s infrastructure head, Santosh Janardhan, highlighted the shift toward evaluating “the whole data center as one integrated system,” including servers, networking, cooling, and power, rather than individual components.

AMD is positioned across multiple parts of that stack, supplying CPUs, GPUs, networking, and FPGAs through its own portfolio, while supporting Meta’s rack architecture.

Meta has already deployed millions of EPYC CPUs and serves as a key validator for Venice.

Anthropic — up to 2 GW, $30.0B forecast below (July 22)

The agreement includes AMD’s commitment to invest up to $5B in Anthropic, with the first GW expected to begin deployment in 1H 2027.

Unlike OpenAI and Meta, Anthropic did not receive a warrant.

Anthropic also demonstrated Claude being used to optimize workloads on AMD MI355X accelerators, with one engineer reportedly improving frontier model performance over a weekend, as mentioned previously in this writeup.

The significance is that the relationship appears to extend beyond procurement into engineering collaboration, while the result was achieved on current-generation hardware rather than the future MI400 series.

Microsoft / Azure — $6.0B forecast below (July 20)

Microsoft is expected to deploy Helios at scale beginning in 2H 2026 for model inference, alongside the new Venice product.

I called out months ago that a Microsoft Helios announcement would likely appear around this event.

In the forecast, I sized the partnership from the language of “multi-billion-dollar” at $6B, which is a conservative estimate with potential upside revision, in my opinion.

SemiAnalysis believes OpenAI may be the anchor customer for Azure’s AMD capacity, representing an additional channel for the broader 6 GW deployment.

Oracle OCI — $1.2B forecast below (October 2025).

Oracle announced plans for 50,000 MI450-class GPUs beginning in Q3 2026.

While this is modest in revenue contribution, the agreement provides a tangible near-term deployment milestone and could become one of the first publicly verifiable Helios installations.

Cerebras — $750M forecast below (July 23)

First, Cerebras and AMD target different parts of the inference workload. Helios handles high-volume processing, while Cerebras’ wafer-scale systems focus on final token generation, where latency is most important.

The companies claim the combined architecture can deliver up to 5x more tokens per second per watt.

The approach is particularly relevant for AI agents, where multi-step reasoning causes latency to compound across each action.

Cerebras will deploy Helios in its own data centers, with the joint offering launching through Cerebras Cloud in 2H 2026.

My $750M estimate is immaterial against a projected $263B of 2030 revenue, but could prove conservative if this two-system architecture becomes a broader industry standard for AI agent workloads.

Financial Projections

Revenue:

The revenue build splits revenue into the contracted GW partnership revenue from the schedule above, server CPUs, other Instinct and data center GPU sales (HUMAIN, DOE, neoclouds, enterprise), and client, gaming and embedded.

Notice how the CPU revenue opportunity is larger than both the other GPU revenue and client / gaming / embedded revenue line items combined.

This may differ from how AMD is usually discussed, but the CPU market is growing rapidly with the pace of innovation in agentic compute.

The server CPU line assumes a market expanding from $25B in 2025 to $220B in 2030, with AMD’s revenue share moving from 46.2% to 50%, in accordance with Lisa Su’s latest commentary during the Summit.

This is a baseline view on the CPU from Lisa Su directly, and could even be underestimated a year from now.

Relative to consensus estimates of $49.7B revenue and $7.48 EPS for 2026, $77.5B and $13.50 for 2027, and $104B and $18.70 for 2028, my model is 7% above consensus revenue in 2026 while essentially in line on EPS ($7.46 vs. $7.48). The model is 14% above consensus revenue in 2027 and 29% above in 2028.

In addition, I am assuming $15B per GW deployed, lining up with the conservative end of Street expectations . $ / GW could easily reach $20B per GW in a bull case.

Consensus estimates also vary widely on the street, with FY2028 EPS ranging from $13.15 to $28.74, highlighting the uncertainty around longer-term forecasts.

Consensus itself has also moved meaningfully recently, rising from roughly $6.15 in early July to $7.48 today.

The 2026 revenue cadence is one of the more testable parts of the model because half the year has already been reported or guided:

• Q1: $10.25B reported (+38% YoY), above the high end of guidance

• Q2: $11.20B (guidance midpoint), reporting August 4

• Q3: $14.50B (my estimate), reflecting initial Helios shipments beginning in September

• Q4: $17.20B, calculated as the amount required to reach the annual forecast

This results in H1 revenue of $21.45B and requires H2 revenue of $31.70B, or ~1.48x the first half of the year, with the majority of the acceleration occurring in Q4 as the MI400+Helios ramps.

Margins:

On margins, the assumptions are more modest than the revenue forecast.

The model uses a 23.0% non-GAAP net margin for 2026, compared with an implied 22.8% in the first half based on Q1 results and Q2 guidance. This requires approximately 23.2% in the second half, implying little change over the first half of the year.

Margins then expand to 28% in 2027 and 31% by 2030, driven by higher rack-scale pricing, server CPU pricing and operating leverage, partially offset by higher memory costs and the lower margins associated with rack-scale systems.

Note: Rack-scale systems have lower profit margins than standalone chips because they bundle semiconductors with expensive, lower-margin physical infrastructure like cooling, power supplies and cabling.

For reference, AMD’s 19.7% non-GAAP net margin in FY2025 was already a company record, so the model assumes continued margin expansion over the following five years.

A couple things worth noting before going any further are two assumptions made regarding timing:

First, deployment timing is intentionally conservative.

The model includes 12.85 of the 14.65 committed GW within the forecast period, leaving roughly 1.8 GW beyond 2030.

Other models out in the market estimate the majority of the GWs can be deployed throughout 2027 and into 2028, which appears aggressive to me given current packaging and memory supply constraints.

Slowing this deployment schedule changes the timing of revenue rather than the total opportunity.

Second, revenue recognition occurs before deployment.

Revenue is recognized when AMD ships hardware to the rack builder rather than when systems are deployed at the customer site (per ASC 606).

Because shipments can occur one or more quarters before deployment, a model based on deployed GW may understate near-term reported revenue, with revenue recognition potentially occurring earlier than deployment milestones.

For comparison, AMD’s long-term targets from its November 2025 Analyst Day included a revenue CAGR above 35%, data center revenue reaching $100B, AI revenue growth above 80%, and more than $20 of EPS within three to five years.

Under my base case, EPS approaches $20 on an exit run-rate basis in late 2027 and exceeds that level on a full-year basis in 2028.

Valuation & Sensitivities

At approximately $530, AMD trades at 70x my 2026 EPS, 34x 2027, 22x 2028, and about 11x 2030, and at roughly 16x P/S.

Those multiples are consistent with a company growing revenue ~50% for five years.

Whether those multiples prove reasonable depends primarily on the earnings trajectory.

My valuation framework assumes multiples compress gradually over time.

The two valuation approaches produce similar outcomes.

P/E implies a 2030 share price of $1,554-$1,776, while P/S implies $1,504-$1,719.

Under both valuation methods, they imply a 28.5-29.4% CAGR.

The valuation framework assumes the earnings multiple compresses from 75-90x in 2026 to 35-40x by 2030. The model does not assume AMD reaches Nvidia’s profitability or software economics, which for reference, Nvidia currently commands TTM net margins of ~63%… I am projecting AMD to mature at 31% net margins in 2030. I’d say under half of Nvidia today is conservative.

To be as accurate as possible, I am assuming 1.638B of DSO (diluted shares outstanding) through 2027, increasing to 1.838B beginning in 2028 to reflect the ~200M shares from the OpenAI and Meta warrants that will dilute the share base.

Because the warrants vest over time, actual dilution would occur gradually rather than all at once. All 2028-2030 EPS figures are fully diluted.

Because a sensitivity grid is difficult to interpret without knowing what each cell represents, the following table maps 2030 EPS against exit multiple, with the EPS rows corresponding to the five scenarios:

To make the sensitivity tables mean something, here’s what each driver is actually worth in share price, using the 37.5x midpoint from the base case forecast:

• One GW of 2030 deployment is worth about $95 a share:

  The path: $15B of revenue, about $4.65B of net income at a 31% margin and $2.53 of EPS. This is the biggest lever in the model. The company has confirmed the September 2026 first shipment and Anthropic’s first GW in 1H 2027. Every other YoY GW figure is based on my personal projections.

• A point of server CPU share is worth about $14 a share:

  On a $220B market, $2.2B of revenue and $0.37 of EPS.

• 1.00% of 2030 net margin: $1.43 of EPS and worth about $54 a share.

• Five turns of the exit multiple: worth about $222 a share.

  Notice that’s only about two GW’s worth. People will argue endlessly about whether 35x or 40x is the right 2030 multiple, and the answer matters less than whether two racks’ worth of GW ship on time.

Risks

I figured I’d put this towards the end to remind the reader not to get too excited on the financial projections. There is still notable risk in AMD’s execution going forward.

Helios Execution:

Helios is the most complex AI rack system launched to date, combining new hardware, networking and OEM partners at significant scale.

SemiAnalysis expects volume production to slip toward Q2 2027, while AMD continues to target its existing timeline.

The key data points to monitor are Q3 and Q4 results.

If Q3 or Q4 revenue is materially below the cadence outlined above, or if they announce a Helios deployment delay in 2027, AMD will sell off aggressively.

Developer Access & Software Ecosystem:

The software progress discussed earlier depends in part on developers having access to AMD hardware.

One challenge internally is limited availability of stable GPU clusters for software development and automated testing, which can slow both software development and the AI coding agents AMD is increasingly relying on. Externally, on-demand access to AMD accelerators also remains limited.

The founder of one of the few cloud providers offering AMD GPUs on demand has argued that AMD has historically underinvested in its cloud ecosystem and that access to MI455X may require multi-year commitments given current supply constraints.

If developers have limited access to current-generation hardware, software improvements may progress more slowly than expected.

Valuation & Expectations:

At roughly 70x 2026E EPS, the current valuation likely reflects a portion of the expected 2027 earnings ramp.

AMD shares DID decline following the Advancing AI Summit, but that is consistent with the stock’s historical tendency to sell off after major product announcements for some odd reason.

If the market tightens, I would view the ~$450 range as a potential area of support.

Nvidia:

Nvidia is expected to bring its Rubin platform to market first, and CUDA remains the leading software ecosystem for AI training.

Nvidia also retains architectural advantages, including its memory-sharing technology, while operating at gross margins above 70% and can provide flexibility on pricing if competitive pressure increases.

Supply, Power & Memory:

Advanced packaging capacity remains constrained through 2026, with AMD controlling a smaller share than Nvidia.

Next-generation HBM memory is also supply constrained, and higher memory costs remain a potential margin headwind.

Beyond 2028, power availability may become a larger constraint than semiconductor supply, which has been a shadowing overhang in compute conversations.

AMD has consistently noted that its long-term market opportunity depends on the availability of power, supply, and capital, which remains an important consideration.

China:

MI308 sales currently operate under export licenses and a 15% revenue-sharing arrangement.

AMD has not disclosed the level of China revenue included in its Q2 guidance, and prior export restrictions reduced annual revenue by approximately $1.5B.

As a result, I treat China as potential upside to the model rather than incorporating it into the base case.

Concluding Thoughts

The central question in my November piece was whether AMD would become a credible second source for AI compute. I believed it, but it wasn’t evident yet to investors.

That question is now resolved, with ~15GW under contract across the eight most significant AI labs, the first full-stack rack shipping in September, the software gap narrowing and server CPUs growing more rapidly than ever.

Execution risk does remain against this contracted demand, and unfortunately, that responsibility does not lie in AMD’s hands, but rather downstream in the supply chain.

My base case requires the MI400+Helios reaching volume between now and mid-2027, GWs cleanly converting near $15B or more each on the ~15GW of negotiated compute, CPU share crossing 50% during or before 2030, net margins reaching 30%+ by EOY 2030 and continuous, successful innovation down the line.

If this holds, AMD can easily reach $1,554-$1,778 by 2030.

Overall, I have never been more bullish on the business.

This is mile 9 of the marathon.

Further reading from event week: Lisa Su’s full CNBC interview and the Helios throughput clip; Su with Brian Sozzi on AI demand and hyperscaler spending; CNBC’s first look inside a Helios rack; Su’s post-keynote press conference on open source and the inference mix; Forrest Norrod on Helios economics; Meta’s Omar Baldonado on open networking; Boppana on ROCm.ai; and the opposing view from SemiAnalysis, which is worth reading particularly if you are long.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

I pitched the stock in a college finance club when it traded around $7.

And I was sitting in a senior-year finance class when the Pepsi partnership hit the news in August 2022.

This piece is the full breakdown of the company as it stands today.

Over the past 18 months, Celsius went from a single-brand growth story to a 3-brand platform holding roughly 21% of the U.S. energy drink category — the #3 portfolio energy drink company in the market behind Red Bull and Monster.

Over that same stretch, the stock fell roughly 40% YTD to around ~$30, well below its 2024 all-time high near $100.

This business transition is very misunderstood by the investing community.

And honestly, this transition screens terribly. A core brand growing 6%, gross margin down ~4%, and a distributor at ~60% of revenue all read as red flags.

I believe every one of those numbers is a snapshot rather than a true representation of what the company will look like in the 2H of this year and into 2027.

Transitions are where the market misprices businesses.

Let’s get into it.

Investment Contents

1. Company Background & Overview

2. Industry Overview & Market Positioning

3. Competition & Consolidation

4. The Pepsi Partnership

5. Alternative KPI Data: Velocity & Distribution

6. Key Areas for Growth

7. Risks

8. Valuation, Financial Forecast & Scenarios

9. What I’m Doing With My Money / Concluding Thoughts

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

Company Background & Overview

Celsius was founded in 2004 in Boca Raton and spent its first decade going nowhere.

The original positioning was a “negative-calorie” diet drink, which struggled significantly with distribution and consumer adoption.

In 2011, the company's stock price plummeted so low that it was delisted from the NASDAQ and forced to trade on the OTC pink sheets. The brand lost roughly half its retail placement and revenue almost overnight as major grocery chains wiped them from the shelves. The company nearly went under.

Instead of filing for court-restructured bankruptcy, Celsius was saved through an out-of-court restructuring and private capital injection.

They were bankrolled by an investor who believed in the business, and Celsius continued to operate as a public company.

John Fieldly joined as CFO in 2012 and became CEO in 2018. Under him, the company repositioned entirely. Revenue went from roughly $52M in 2018 to $1.32B in 2023.

The repositioning worked because it found white space. Traditional energy (Monster, Red Bull, Rockstar) sold sugar and adrenaline to a roughly 70/30 male-skewed customer. Pre-workout powders sold intensity to gym-goers.

Celsius positioned between them: a functional energy drink with a proprietary ‘MetaPlus’ blend, backed by university clinical studies, a clean white can with fruit imagery, and explicit zero sugar, no aspartame callouts that made it feel safe for daily consumption.

The customer base that showed up looked nothing like the traditional energy consumer. Celsius sells to a nearly 50/50 gender split and gears toward $100K+ household incomes.

Distribution did not start in mainstream beverage aisles early on, but rather in coolers in CrossFit boxes, boutique gyms, GNC and Vitamin Shoppe. When Celsius re-entered grocery, it negotiated placement in health and wellness areas instead of the crowded energy aisle.

By the time the rest of the better-for-you (BFY) wave scaled up, Celsius held the fitness demographic and the shelf space around it.

The marketing was built on the same idea — to meet the customer where they already are.

Instead of extreme sports sponsorships, Celsius made “Live. Fit. Go.” a lifestyle slogan. Fieldly has said he wanted the can to work like the Starbucks “green straw” — a visual signal that the person holding it takes wellness seriously.

The company used everyday personal trainers and wellness creators on TikTok and Instagram rather than superstar athletes, ran branded run clubs through Strava and showed up at lifestyle events like the Breakaway music festival (EDM festival full of the exact 18 to 26-year-old consumer Celsius targets).

Today that has scaled into Formula 1 (Aston Martin Aramco), MLS, and Formula DRIFT sponsorships.

Even though Celsius seemed like an overnight success, it was not. After 2 decades of brand building, during 2025, the company yielded:

• ~$5.2B in annual U.S. retail sales across the portfolio

• Sold in more than 250,000 tracked U.S. outlets at 99.5% ACV

• Household penetration of 34% for the core brand, 22% for Alani Nu, and roughly 43% for the portfolio

• Repeat purchase rates above 65%

For context, Monster needed 8+ years to reach 40% U.S. household penetration. Celsius reached 34% in roughly three years of scaled distribution.

Fieldly framed the current strategy as:

“We’re building a portfolio that reaches more consumers during more occasions, and we’re doing it with discipline, collaboration, and a commitment to organizational excellence.”

The management team was rebuilt around that goal, with Eric Hanson as President & COO, Jarrod Langhans as CFO, a new CMO, and a dedicated President of Celsius International.

Industry Overview and Market Positioning

Energy drinks are the growth engine of the entire non-alcoholic beverage industry right now.

U.S. energy drink sales reached approximately $24.8B for the 52 weeks ended in early March 2026, up almost 14% YoY, which is the fastest growth of any major packaged beverage category.

Energy is now the second-largest non-alcoholic packaged beverage category in the country, behind only regular carbonated soft drinks at around $30B.

The category is also concentrated. The top five companies control over 90% of U.S. energy drink sales, and within that group, Celsius and Keurig Dr Pepper grew more than 20% over the trailing year while the incumbents grew single digits. Industry forecasts have the category compounding around 8% annually over the next decade.

More importantly, 85% of 2025’s energy drink growth came from zero-sugar products.

And the category is splitting.

Celsius management says the two halves are traditional energy (transactional, sugar-heavy, impulse-driven) and modern energy (zero sugar, daily routine, lifestyle and wellness).

This maps onto the broader move toward healthier food and drink in the U.S.

More than 85% of American adults consume caffeine daily, which coffee still accounts for roughly 60% of those occasions, but energy drinks are continually converting more and more young adults to energy drinks.

More than 2 in 5 Gen Z consumers drink energy drinks regularly, double the rate of millennials and older cohorts.

The two halves of the category also market differently. Traditional energy built its equity on extreme sports, gaming, and truck-stop culture (think Monster). Modern energy markets at fitness events, run clubs, music festivals, and campuses (think Celsius).

Celsius sits at the center of that shift: the portfolio accounted for 33% of all zero-sugar RTD energy growth in 2025, the largest single contribution in the category.

Regarding trend durability, roughly 32% of consumers are drinking energy drinks on more occasions than before and 51% of category growth comes from increased frequency among existing buyers (not new users trying the category).

In addition, 37% of Celsius consumers drink it with a meal, and 33% drink it socially.

The female consumers, the demographic Celsius and Alani Nu index toward, consume energy earlier in the day and as part of a routine, versus the impulse-driven male consumer traditional energy was built on.

—> Routine consumption means repeat purchase. Repeat purchase is what retailers give shelf space to.

Not every challenger captures this. Prime, the viral 2022-23 brand founded by Logan Paul and KSI, saw sales fall nearly 65% over the past year. Monster’s acquired brands are also shrinking, with Reign down 7%, Reign Storm down 4% and Bang down over 4%.

Brand image/loyalty and product velocity support long-term growth of brands in this category.

International

Celsius still generates roughly 95% of revenue from North America.

Monster generates approximately 45% of its sales outside the U.S.

That gap between the two companies is either a hole in the Celsius story or the biggest untapped growth lever, depending on execution from here.

Run the simple math — if Celsius ever reached a Monster-like international mix on today's revenue base, international would be a $2B+ business. Today it is around $93M.

The numbers are small for Celsius but the segment is directionally moving well, as international revenue reached $35.3M in Q1 2026, up 55% YoY, with Europe at $25.4M (+36%) and Asia-Pacific growing over +200%.

Some investors flagged that international actually shrank as a share of total revenue (6.9% to 4.5%) — but that's just the denominator, not the segment. The U.S. business is simply growing that much faster with Alani still on its growth curve.

When I was looking deeper in the international footprint, I was surprised, as it’s wider than most realize. Celsius has presence in the UK, Ireland, France, Australia, New Zealand, Benelux, and the Nordics, and entered Spain in March 2026 through an exclusive distribution agreement with Suntory Beverage & Food, with Portugal next under the same partnership.

The whole operation is now coordinated from a global headquarters in Dublin as well, under a dedicated President of Celsius International, which was a role created specifically to run this expansion. The man in charge of International is Garrett Quigley, who actually previously spent 25+ years with Pepsi, most recently as SVP of Global Franchise.

This is a great sign for Celsius, as they are bringing in someone with robust experience from their closest distribution partnership.

Separately, regarding the international strategy of Celsius, they are not focused on building their own infrastructure country by country. Rather, Celsius signs established local operators who already own the routes, the retail relationships, and the reps.

Fieldly describes the approach as “key markets, strong local partnerships, disciplined launch plans and sustained marketing and distribution support.”

It is the same capital-light playbook Coke and Monster used to take energy global.

There’s some early evidence the brand is traveling well abroad as well, with Celsius reaching 88% grocery penetration in the UK and France within 12 months of entry.

This is a great sign of progress.

The energy category penetration across most of Europe DOES trail the U.S. by years, with the zero-sugar modern segment barely existing abroad yet, and the health-and-wellness positioning maps onto markets like the UK, Australia, and Benelux where that consumer trend is already established.

It won’t map to every market due to differing consumer habits and regulatory hurdles, but Celsius has great momentum

Competition & Consolidation

Red Bull holds roughly 35.9% of the U.S. category.

It sold a record 14B cans globally in 2025 (+10.2% YoY) with global sales up 8.6% to over $14B, and its Red Bull Zero line grew nearly 200% following its early-2025 wide release. Red Bull clearly also realizes the zero-sugar shift is where the growth is.

Red Bull is also the category’s structural outlier, as it invented modern energy drinks and built its own global distribution network, something no brand since has replicated.

Monster is the next big dawg and holds roughly 27.3%. It posted its first $2B quarter in Q4 2025 (net sales +17.6%) and grew Q1 2026 net sales 26.9%, which is actually a genuine reacceleration after a soft 2024.

The core Monster Energy segment still produces about 93% of its revenue, distributed globally through Coke’s system since 2015, and the stock hit an all-time high in July 2026 at a market cap around $92-96B. Notably, Monster’s growth is coming from its namesake and Ultra families, while its acquired brands (Reign, Bang) are shrinking.

So the incumbents are healthy and reaccelerating growth.

This makes the data interesting. The Celsius portfolio reached 20.9% of the category in Q1 2026, and legacy Celsius is, to date, the only brand in the category’s history to take meaningful, sustained share while Red Bull and Monster were both still growing.

Prior challengers (Bang most famously) took share briefly and collapsed.

Note: Image below is outdated and excludes Alani Nu, but includes Rockstar

The category now functions as a three-portfolio market:

• Red Bull

• Monster

• Celsius

Where the three companies differ most is the consumer they're built around.

Red Bull and Monster built their equity on extreme sports, gaming and high-octane masculinity, and their heaviest users skew male and is impulse-driven.

The Celsius portfolio was built around fitness, wellness and a 50/50 gender split.

All three are now converging on the same zero-sugar consumer — Red Bull with Zero and Monster with Ultra, which means the modern energy segment Celsius pioneered is no longer uncontested and competition has intensified.

The incumbents bring more distribution muscle and marketing budget to this fight, but Celsius brings the brand equity of having defined the space.

New Entrants

The BFY positioning became a more fierce playground for Celsius around 2021.

Ghost, Zoa, Alani Nu, and a long tail of founder-led startups all entered the energy segment, each with a niche:

• Ghost through flavor collaborations and gaming culture

• C4 through supplement stores and Bloom Nutrition through traditional channels

• Zoa through Dwayne Johnson’s fitness following

• Alani Nu with the young female consumer.

We’ve seen consolidation since in the past few years.

Ghost sold 60% majority ownership to Keurig Dr Pepper for $990M in late 2024, at an implied valuation around $1.65B, with a path to full KDP ownership by 2028.

Zoa was bought by Molson Coors for $53M in cash in late 2024.

Bang (not BFY, but worth mentioning) held roughly 10% of the category at its 2019 peak, but went through Chapter 11 bankruptcy and was bought by Monster for $362M in 2023. It has kept shrinking inside Monster’s portfolio.

Celsius ended up on the other side of that table and was the acquirer.

Alani Nu

Celsius closed the $1.8B acquisition of Alani Nu in Apr-25. The price worked out to under 3x net revenue and roughly 12x fully synergized EBITDA with minimal leverage.

The results since close:

• Approximately $1B of revenue contribution in 2025, and roughly $1.6B in retail sales over the trailing year — nearly double YoY

• $368.1M in Q1 2026 net sales, up approximately 60% PF, with scanner growth near 100%

• ACV up from the high-80s to 94.2% after moving into Pepsi’s DSD network

• 9.0% share of the U.S. category on its own

On a run-rate basis, Alani Nu has already passed the legacy Celsius brand in size.

Why did this work so well for Celsius? A few things:

1. First, energy is the lifeblood of Celsius and they have been operating for 20+ years.

2. Second, is the demographic — Alani owns the young female wellness consumer the way legacy Celsius owns the fitness consumer. The two brands overlap less than you would think, causing less cannibalization. I’m not sure Celsius would have made this acquisition if cannibalization was the case.

3. And third is the continuous innovation — Alani runs rotating limited-time flavors (Cherry Bomb in Q4, Lime Slush in Q1, each becoming the brand’s top seller in tracked channels), which keeps trial and social media attention refreshed.

What’s crazy is the awareness data illustrates there is tremendous remaining runway for the brand — Alani sits at only about 20% brand awareness nationally. Among consumers who know it, purchase conversion runs around 72%, which is comparable to Red Bull. Low awareness with high conversion paired with a recent acquisition where the company has dominant domestic distribution and international exposure is a perfect combination. The constraint here is visibility, not the product, and Celsius/Pepsi’s network will help amplify the visibility of the brand.

Rockstar

In Aug-25, Celsius acquired the U.S. and Canadian rights to Rockstar Energy from Pepsi for $585M in convertible preferred stock — this transaction lifted Pepsi’s equity stake to 11%, gave Pepsi a second board seat, and made Celsius Pepsi’s energy captain in North America.

Rockstar is the weakest asset in the portfolio and still has to be fully integrated into Pepsi’s DSD network. Q1 2026 revenue was $66.6M with retail sales down (13%) YoY. The brand held flat net shelf space through the spring resets.

Fieldly has described 2026 as a stabilization year for Rockstar, to focus on fixing the SKU set, sharpen positioning and improve execution, with a rebuild he has framed as taking “a handful of years.”

So why’d Celsius buy this brand if it was going to be so much work?

The logic for buying it is that Rockstar covers a portion of the market where legacy Celsius and Alani don’t. This is the convenience and gas channel, the value price tier, and the traditional male energy consumer, which is a core energy market worth around $17.8B that skews 71% male.

With all three brands, Celsius can offer a retailer a complete shelf solution for every energy segment. Neither Monster nor Red Bull can make that claim across modern and traditional energy.

The Pepsi Partnership

The Pepsi relationship is the most important fact about Celsius, and the most misunderstood, in my opinion. It deserves the full history, including the part that went awry a couple years ago.

The 2022 deal

In August 2022, Pepsi invested $550M for an 8.5% stake and took over long-term distribution.

The immediate unlock was access to Celsius’ core target market. Pepsi held exclusive beverage contracts covering campuses with roughly 61% of the U.S. college student population, which was a channel Celsius had struggled to reach historically. This also included the vending, foodservice, and convenience distribution capabilities of a top-two global distributor.

Celsius ACV climbed toward the high-90s over the following two years, and revenue grew 102% in 2023.

The revenue recognition problem

Celsius recognizes revenue when product ships to Pepsi, not when a consumer scans and buys a can.

After the 2022 deal, Pepsi loaded inventory aggressively to avoid shortages. That means a meaningful piece of 2023’s spectacular reported growth was Pepsi’s warehouses filling, not consumers actually purchasing that inventory.

By early 2024, Pepsi had stocked an estimated $100M-$120M, which was more than consumers demanded, and the unwind ran through Celsius’ P&L for around a year.

One stretch of late 2024 saw reported revenue take a 31% hit while retail sell-through remained stable and growing. The stock fell 52% in 2024 as investors read the shipment decline as brand decay.

—> The growth on the way up was overstated, and the decline on the way down was overstated, by the same accounting policy of revenue timing.

Separately, Celsius paid a $3M SEC penalty and consented to a cease-and-desist over materially inaccurate 2021 financial statements tied to accounting-control failures around stock awards. A 2026 class action alleging executives misrepresented the Pepsi inventory dynamics was dismissed quickly.

So a practical lesson for anyone analyzing this company is that reported revenue and consumer demand routinely diverge around distribution transitions. It’s definitely an annoying financial aspect of the business, but I think this helps contribute to the divergence of opportunity and asymmetric upside.

The 2025 restructuring

In August 2025 the alliance deepened between Pepsi and Celsius, when it was announced that Alani Nu is moving into Pepsi’s DSD network, Celsius bought Rockstar’s domestic rights, and Pepsi’s equity stake rose to ~11% in Celsius. It was also announced then that Celsius formally took over the energy category for Pepsi design, SKU prioritization and promotional execution in the U.S..

By Q1 2026, Pepsi accounted for 59.0% of Celsius revenue.

That concentration number is the first thing analyst skeptics raise, so it’s worth laying out why the structure exists and why the history of this category suggests it can be an advantage rather than a risk.

Coke bought ~16.7% of Monster in 2015 for $2.15B and became its preferred global distribution partner across 190+ countries. That transition was painful, with Monster’s Q1 2015 net income falling 95% on distributor termination costs, with full-year termination obligations of $218.2M.

Following this pain, the core Monster brand compounded domestically and internationally utilizing Coke’s distribution model into a company worth roughly $92-96B today.

Celsius has been running that same sequence. Approximately $400M in distributor termination fees were realized through its P&L during the transition (funded by Pepsi) — this is the same short-term earnings distortion and resulting market skepticism. The remaining termination balance fell from $264M in December to $40M by the end of Q1.

The incentive structure is also different from a normal vendor or customer relationship. Pepsi owns 11% of Celsius, holds two board seats, and named Celsius its energy captain after its own energy brands (Amp, Kickstart) shrank 28% over the past year to around $130M.

Pepsi spent 15 years trying to build energy internally before handing the category to Celsius.

More broadly in this category, every major energy brand utilizes a conglomerate’s distribution:

• Monster with Coke

• Ghost with Keurig Dr. Pepper

• C4 with Keurig Dr. Pepper

• Celsius with Pepsi

• Zoa with Molson Coors

Red Bull is the only self-distributing exception, and it invented the category with a two-decade head start.

Whether the ~60% Celsius/Pepsi concentration reads as a moat or a vulnerability is a judgment call — I’ll give you mine in the risks section below, but the history section above is what I’d weigh it against.

Alternative KPI Data: Velocity and Distribution

“In retail, if you are off-shelf, you are not selling,

We live or die each day by our supply chains.”

— John Fieldly, CEO

Rather than rely on the reported financials, this section goes around them and focuses on distributor disclosures, distribution KPIs, retail data, ecommerce rankings, consumer survey work, and the company’s job postings.

Shipments vs. Scanning

This table stacks the past four reporting periods of core-brand wholesale revenue against U.S. retail scanner demand:

During four straight periods where reported revenue and consumer demand pointed in different directions, consumer sell-through of the core brand stayed positive and stable while shipments moved with Pepsi’s warehouse decisions.

By Q1 2026 the order patterns had normalized: record consolidated revenue of $782.6M, the full $50M Alani synergy target captured, and the distributor termination cleanup nearly complete.

The 6% legacy Celsius brand growth that spooked investors deserves a similar perspective — Celsius deliberately cut its slow-moving flavors ahead of the spring resets.

These cuts happen instantly while the replacement shelf space builds over months.

As COO Eric Hanson put it on the Q1 call, “we see the reduction faster than the ACV build.”

Dollars per point of distribution rose while total distribution points dipped, which is a sign of a productivity tradeoff rather than demand decay.

Management has indicated that Q2 should be a side-step in settled demand, with ramp turning around during the end of the year and into 2027.

This is definitely one of my closest watch items as we head into the remainder of the year.

Shelf Expansion & Optimization

The expansion of Celsius from a single-brand upstart into a multi-brand energy platform has reshaped its physical retail footprint and category placement.

During the spring 2026 category resets, which finalized between June and July, Celsius successfully expanded its physical shelf space across grocery, convenience, mass, and club channels:

• Core Celsius: +17% incremental shelf space across national chains, including new cooler doors and secondary placements

• Alani Nu: ACV distribution rose from the high-80% range in late 2025 to 94.2% by early 2026, driving +100% space gains across channels

• Rockstar: flat net space while the SKU set gets rebuilt

To complement its presence in traditional retail channels, Celsius has expanded into foodservice, which now accounts for ~11% of the company’s Pepsi-routed revenues.

Key milestones in this expansion include national partnerships with Dunkin’ and Jersey Mike’s Subs. Dunkin’ added Celsius Sparkling Orange to its beverage coolers, while Jersey Mike’s introduced a three-flavor lineup nationwide.

However, this rapid retail expansion faces headwinds from private-label products.

In early 2026, Costco launched its Kirkland Signature Sparkling Energy Drink, priced at $16.99 for a 24-can variety pack containing Peach, Orange, and Tropical flavors.

This private-label product is positioned as a direct competitor to Celsius, matching its 200mg caffeine dose, zero-sugar formulation, and key vitamins at a 40% to 55% discount to the branded equivalent.

Because Costco represents approximately 11% of Celsius’ total sales, the introduction of this private-label alternative caused a temporary 7% decline in Celsius stock and presents a risk of velocity dilution in high-volume club store Costco channels.

Despite this, Celsius has a much stronger brand image and marketing strategy / scheme, and pop-up brands like this Kirkland Signature beverage fade quickly once after consumers try it a couple times.

Digital velocity

While physical shelf expansion remains the primary driver of bulk volume, digital e-commerce channels serve as a key leading indicator of consumer pull and velocity.

E-commerce demand for functional energy beverages remains strong, with the online retail energy drink segment projected to grow at a 7.5% CAGR through 2035.

Within this digital landscape, Celsius maintains a leading market share position, routinely outpacing legacy incumbents on digital platforms.

On Amazon, Celsius remains the top-selling ready-to-drink (RTD) energy drink brand. During peak promotional periods, such as Amazon Prime Day, Celsius achieved an 18.4% weekly market share in the energy drink category (2025 Prime day).

Right now as I write this, it appears Celsius still appears at #1, and also appears at #4, #7, #18 and in other places, with Alani starting at #15 and #16. Rockstar starts appearing at #38, which is pretty far down the list, but matches up with management’s commentary and intention of turning around this business.

This digital velocity is driven by the performance of multi-flavor variety packs, which attract consumers looking for convenient home delivery and bulk savings.

According to digital influencer indexes and e-commerce tracking data, the Celsius Assorted Flavors Official 12-Pack Variety (featuring Sparkling Orange, Peach Mango Green Tea, Kiwi Guava, and Fuji Apple Pear) holds a prominent position within e-commerce search rankings, placing at #46 in the highly competitive Grocery & Gourmet Food category influencer index.

This strong e-commerce performance is supported by high customer acquisition and conversion rates. Analysis of consumer purchasing behavior indicates that Celsius displays high conversion efficiency relative to its total brand awareness.

Awareness vs. conversion

The consumer survey data reframes the competitive picture:

• Red Bull: 82% brand awareness

• Monster: 75%

• Celsius: 49% — with 63% of aware consumers connecting the brand to an active purchase occasion

• Alani Nu: 20% awareness, but converting around 72%

Half the country still hasn’t properly met these brands yet, while the consumers who have met them buy at rates comparable to the incumbents.

That combination of low awareness and high conversion is what distribution systems and marketing budgets exist to exploit, and Celsius now has access to both at Pepsi scale.

Additionally, Celsius’s digital consumer base has a favorable demographic profile, showing high penetration among $100K+ high-income households.

—> This demographic is relatively insulated from inflationary pressures, supporting premium price resilience on digital platforms.

Organizational Capability & Field Sales Hiring

To support Celsius’ retail space gains and ensure execution at the shelf, Celsius has scaled its internal sales organization.

They’ve transitioned their field operations from a reliance on external agencies to an in-house model while prioritizing execution quality to monitor SKU compliance, display compliance and local distributor alignment.

An analysis of Celsius Holdings’ active job openings in 2026 reveals two primary operational profiles that drive regional execution:

• Field Execution Representatives (FERs): This entry-level role is paid $25 per hour plus overtime and a vehicle allowance. Operating on a Wednesday-to-Sunday schedule, FERs are tasked with setting up in-store displays, securing end-cap placements, conducting product sampling demos, and executing 15 to 20 account audits per day to ensure correct pricing and shelf alignment.

• Territory Sales Managers (TSMs): This professional-level role features a base salary range of $60,000 to $65,000 per year plus performance bonuses and RSUs. TSMs manage regional distributor partnerships (such as local Pepsi bottlers), conduct distributor “route rides” and “market blitzes,” and oversee 12 to 15 store visits per day to resolve shelf distribution gaps.

The geographical distribution of these open positions reveals the focus areas of Celsius’ retail strategy.

Rather than focusing solely on mature, high-volume states, the company’s hiring activity is concentrated in regional clusters designed to expand distribution depth and defend market share in highly competitive markets.

This hiring pattern demonstrates that Celsius is actively allocating capital to feet-on-the-street merchandising. This field sales presence is designed to minimize out-of-stock scenarios and secure secondary placements during high-volume seasonal windows.

Key Areas for Growth

There are roughly six areas for growth that I believe will determine the next 12-36 months of Celsius’ stock performance:

1.) Legacy Celsius brand growth

The past year was the first in which this organization managed a multi-brand portfolio, and management’s attention went where the integration risk was.

That work is now largely complete with Alani’s integration finished, synergies captured and the SKU standardization finished.

Fieldly’s argument on the Q1 call was that consistency itself is the growth lever, which means the same core flavors in every store, because “consistency drives repeat purchase.”

The setup for the back half according to management is 17% more shelf space, a permanent Fizz-Free sub-line scaling at rising velocity and a World Cup activation cycle for the core Celsius brand.

The Monster precedent is worth keeping in mind as its flagship Monster drink reaccelerated after the 2015 Coke distribution transition ended. Whether the Celsius flagship does the same is the central question of H2 2026.

2.) Channel whitespace inside Pepsi DSD

Campus contracts covering around 61% of U.S. college students continue to deepen into vending, dining halls, campus markets and peer-marketing programs.

Beyond campuses, hospitals, medical centers, and shift-work environments (think hotel grab-and-go pantries, resort pool bars, and cruise lines).

Cross-CPG plays only a Pepsi partner can run would be a growth driver as well — like meal-deal bundling at checkout and participation in Pepsi’s digital loyalty ecosystem.

Each of these is individually small but together they extend distribution into everywhere Pepsi already does business.

3.) Shelf optimization

ACV (all-commodity volume) measures the percentage of total market sales flowing through stores that carry your product, weighted by store size.

So being in Walmart counts far more than being in a corner shop.

Celsius’s portfolio ACV is 99.5%, which means the product is already in effectively every store that matters. From here, growth comes from more space, better placement and faster sales cycles within existing stores. This is actually what the reset gains and SKU rationalization are engineering!

Alani was the exception, as its ACV was deliberately pulled from the high-80s to 94.2% by plugging it into Pepsi DSD.

The clean way to track all of this would be weekly scanner growth, as NielsenIQ and Circana data is the ground truth on velocity and share, but that data sits behind five-figure institutional paywalls, so the tracked-channel disclosures KPIs explained above are the available proxies.

4.) Bolt-on M&A

This is essentially replicating the Alani / Rockstar playbook — buying a high-growth, independent brand, plug it into Pepsi’s DSD, increase distribution and expand shelf space, and repeat.

The balance sheet has net leverage around 1.0x, which is more than enough room to fund another deal.

I’ve screened for independent, founder-led BFY brands whose formulations fit the portfolio’s clean-energy identity (ex. green tea caffeine, yerba mate, adaptogens), and it produces a reasonably short list.

I’ve put together an M&A deck below (since that’s what I’m good at, lol):

Note: Links are provided below to each brand’s website

Joyburst

Gorgie

CLEAN Cause

Odyssey

Proper Wild

Other brands such as Uptime, Gym Weed, Marquis, and Phocus are further down the list.

None of these is required for the growth story though — M&A is all incremental to my thesis

5.) International expansion & growth

As mentioned previously in the industry section, 95% of Celsius’ exposure is in North America versus Monster’s roughly 45% international mix.

Celsius is executing on a similar strategy to Monster back in the day, and if successful, will boast meaningful growth for Celsius over the years. This includes strong local distribution partners, market-by-market launches and sustained marketing support.

An example of this is the Suntory Beverage & Food agreement in Spain signed in March of this year. Portugal is next under the same distributor, layered on top of the UK, Ireland, France, Australia, New Zealand, Benelux, and the Nordics.

It’s the same international business plan Monster used with Coke’s bottlers

6.) Category adjacency

This one is a long shot but I thought it’d be fun to include — food for thought.

Monster owns alcohol brands, so the question comes up: should Celsius eventually buy a BFY alcohol brand, an electrolyte company, or a functional / nootropic platform?

The Monster experience argues for caution, as it impaired roughly 55% of its CANarchy craft-brewery purchase and the alcohol segment still loses money.

Alcohol also runs through a legally separate three-tier distribution system Pepsi can’t touch, and Monster notably had to hand its non-energy brands to Coca-Cola when their alliance formed (the same would have to be done with Pepsi, or Pepsi wouldn’t allow it under their agreement).

Hydration, functional wellness and nootropics are the more natural adjacencies though, and Celsius already sells hydration powders which are similar.

This is all optionality.

Risks

There are roughly six areas that I believe could impair Celsius’ business and stock performance over the next 12-36 months:

1.) Legacy Celsius brand growth fails to reaccelerate

One of the largest risks is that the core Celsius brand does not return to meaningful growth following the Alani Nu integration.

My base case assumes the recent ~6% growth reflects temporary disruption from integration activities rather than a structural slowdown in consumer demand.

—> For context, this 6% comes after a quarter where the company reported growth of 7.5% for Q4 2025 and 13% YoY growth for Q3 2025… a QoQ step down 3 times in a row, which is why the stock has sold off so violently.

If the core brand remains around this level into early 2027 — after distribution resets are complete and major marketing investments such as the World Cup are behind the company — it would suggest the flagship brand has matured sooner than expected.

And that would cause a continued decline in the stock price most likely.

In that scenario, Celsius increasingly resembles a multi-brand beverage platform rather than a high-growth company anchored by its original brand, which would likely warrant a lower valuation multiple as well.

Scanner data continues to suggest underlying demand remains healthy, but the second half of 2026 will be an important period for validating that assumption.

2.) Commodity costs, tariffs, and slower margin recovery

Gross margin declined to 48.3% in Q1 2026 from 52.3% a year earlier, primarily reflecting higher aluminum costs and tariff-related pressures. Aluminum, the company’s largest packaging input, experienced meaningful price increases during the quarter, and management has indicated Q2 will likely represent a temporary pause in margin improvement.

The company has several initiatives intended to offset these pressures, including locked-in conversion costs extending into future years, a second North Carolina production line expected to come online in the second half of 2026, and continued optimization of its pricing and package mix.

Management expects margins to improve through the remainder of 2026 and into 2027.

But if commodity inflation remains elevated for longer than expected, margin recovery could simply take longer than currently projected.

3.) International adoption of BFY energy drinks develops more slowly than expected

A meaningful portion of the long-term growth opportunity depends on international markets adopting healthier energy drink alternatives at a pace similar to the U.S.

That outcome is not guaranteed, particularly as several markets have introduced or proposed tighter regulations around high-caffeine beverages. The UK has announced restrictions on sales of high-caffeine energy drinks to consumers under 16 beginning in 2027, Canada limits caffeine content to 180mg per serving, and proposed U.S. legislation would require additional warning labels on beverages containing more than 150mg of caffeine.

Current regulations appear manageable, but slower international adoption could delay Celsius’ expected expansion.

Sell-through trends in markets such as the UK and Spain will likely be more informative than announced distribution gains alone.

4.) Regulatory scrutiny around youth marketing

In June, the Texas Attorney General opened an investigation into Alani Nu’s marketing practices following a consumer death.

While the legal merits remain uncertain, the case highlights the broader regulatory attention surrounding energy drink marketing.

Celsius has previously resolved regulatory matters without material long-term financial impact, including settlements related to labeling and SEC disclosures.

The primary concern is less about direct financial liability and more about reputational risk if additional investigations increase consumer scrutiny around energy drink safety, particularly among younger consumers.

At this stage, I view this as a risk worth monitoring rather than a change to the long-term investment thesis.

5.) Competitive intensity increases

Competition within the energy drink category remains significant.

Private-label offerings such as Kirkland continue to expand at meaningful price discounts, while established competitors including Monster and Red Bull continue introducing new products and investing behind their brands. Pepsi and Starbucks are also growing their energy portfolio from a relatively small base.

We also have other brands popping up, similar to Celsius, that are quite popular this summer, such as Bloom Nutrition, a brand of C4, whose parent company is Nutrabolt.

Celsius does retain several competitive advantages, including strong brand recognition, a differentiated portfolio following the Alani acquisition and leading product velocity within many retail channels.

However, shelf space remains highly competitive, promotional activity could increase, and maintaining pricing power may become more challenging if consumers become increasingly value-conscious.

6.) Customer concentration with Pepsi

Pepsi accounted for approximately ~60% of Celsius' revenue in the most recent quarter, making concentration one of the most frequently cited risks.

While the distribution partnership has been instrumental in the company's growth and resembles the model Monster used to scale nationally, it also creates meaningful dependence on a single partner.

Any future contract renewal or renegotiation would occur with Pepsi fully aware of that dynamic, which could likely mean margin pressure for Celsius.

The company's previous transition to Pepsi also resulted in approximately $400M of one-time termination costs, illustrating the financial and operational complexity involved in changing distributors.

Although I currently view the partnership as a net positive given the distribution reach and aligned incentives, the concentration risk remains real and should be considered as part of the overall thesis.

Valuation

At today's price, Celsius trades near the low end of its historical valuation range and grows materially faster than most large beverage peers.

At around $31, CELH trades at approximately 19x forward earnings on consensus FY26 adjusted EPS of around $1.60.

Its five-year median forward P/E is roughly 77x.

Monster, the closest comparable, grows consolidated revenue a fraction as quickly but trades around 37-40x forward earnings, near its five-year average of 35-40x.

On sales, Celsius trades around 2.3x forward revenue against Monster at approximately 10x.

The trailing GAAP P/E for Celsius appears elevated (around 70x on some finance sites) because reported earnings still include one-time distributor termination costs (remember, Pepsi funded this) and acquisition-related expenses. Those charges reduce the usefulness of trailing earnings when comparing today’s valuation to prior periods.

I’m also not assuming Celsius should trade at Monster’s valuation though…

Monster continues to generate higher margins (55%+ gross) and has a substantially longer operating history.

The comparison simply illustrates that Celsius currently trades at a significant discount despite materially higher expected revenue growth.

Applying Monster’s five-year average 35-40x to the Street’s $1.60 gets you $56-64 — right where sell-side consensus already sits ($64-68, Strong Buy) — that’s without relying on the long-term assumptions used later in this model.

For a nearer-term anchor: my own model’s 2026 EPS at a 30-40x multiple implies a fair value range of roughly $38-51 this year.

That’s an implied move of +26-70% before year end, with a midpoint at +48%.

Two things stand out in the sensitivity table above.

The first is which axis drives the most change. Moving down a margin row adds $11-12 to the implied share price, while moving across a growth column adds only $1-2.

This implies that this is a margin story before it’s a growth story.

Whether Celsius grows 8% or 12% matters far less than whether net margins land at 10% or 14%, which is why the forecast section spends most of its time on the margin bridge.

Top-line growth is more of a sentiment bar for investors, which can have a material effect on stock prices, so I wouldn’t be surprised if a beat on the topline drove the stock up materially.

Second, the low end of the table (still referring to the above sensitivity table, not below): at 6% growth and 8% margins, a scenario where the core brand stalls and the margin recovery largely fails, the model still produces $46 at 30x.

That’s still a 50%+ return in a bear case.

The current share price sits below every cell in the grid, which implies the market is either assuming fundamentals below these ranges or applying a multiple well under 30x.

This other sensitivity table above shows where the stock sits today in the top-left corner. Around $30 is almost exactly 20x on $1.50, which is the lowest multiple on the table applied to the lowest EPS on the table.

The Street’s numbers with Monster’s historical 35-40x multiple land in the $56-64 band. This implies a return of +87-113%.

My base case ($3.00 of EPS by 2030) at 30x is $90, which implies a return of 200%.

The table also covers the ‘compression’ case, which is if the multiple never moves past 25x, $2.50 of earnings still produces $63, which implied a return over 100%.

One last way I like to frame this valuation is on a relative basis through market capitalization.

My P/E base case scenarios in my forecast model imply a $23-31B market cap in 2030, roughly 24-33% of what Monster is worth today.

Even under the higher-end P/S scenarios ($41-46B), Celsius would still represent less than half of Monster’s current market capitalization, with 4+ years of category growth baked in.

In other words, the model does not require Celsius to reach Monster’s current scale for the investment to work.

Financial Forecast

This is one of (and always is) the most important sections of the piece, so let’s walk the model line by line.

Every forward number below is my estimate, not guidance from the business.

Revenue:

I’m modeling $3.27B in 2026 (+30%), then +15% in 2027, +12% in 2028, and +10% in 2029 and 2030, reaching approximately $5.10B by 2030, which is a 15.2% 5yr CAGR.

The 2026 number is mostly back of the napkin arithmetic, which assumes:

• Alani annualizes toward its roughly $1.5B retail run-rate with scanner growth still near triple digits

• The core brand comps ease in H2 once the integration reset noise clears

• Rockstar stabilizes

• International grows modestly off a small base

Note that I’m below the Street on revenue (consensus sits near $3.45B for 2026), and the deceleration from 2027 forward assumes Alani matures quickly and no fourth brand gets added.

—> Conservative on purpose.

Margins:

I’m modeling net margins recovering from around 3% in 2025 to 10% in 2026, then growing modestly 1% a year — 12%, 13%, 14%, and 15% by 2030.

A 3% to 10% jump in one year is a big move, I agree, so let me explain where this is coming from.

But honestly, it matches what the Street’s 2026 numbers already imply, which is what makes it a base case rather than a bull case, but obviously still deserves a real explanation, because margins don’t triple on hopium (aka hope):

• FY2025 absorbed approximately $400M in one-time distributor termination fees (expensed through the P&L even though Pepsi funded them), acquisition and integration costs and inventory step-up accounting.

  • Strip these one-timers and the underlying business was already running far above a 3% margin.

  • A meaningful piece of the jump is simply those costs not repeating.

• Net income to common was $85.1M on $782.6M of revenue in Q1, nearly an 11% margin.

  • The 10% full-year assumption is just the business roughly holding what it just reported, through a Q2 aluminum ‘side-step’ and World Cup marketing spend.

• The $50M of Alani Nu synergies is captured in full this period, with consolidated procurement across three brands (aluminum, sweeteners, ingredients), co-packing consolidation, back-office and marketing overhead reduction, and freight and warehouse optimization (although Rockstar still has a bit to go, but this is a much smaller $ share) creating compounding value into 2H 2026 and forward.

• The past year of management attention went to integration — think distributor transitions, SKU reconfiguration and moving over various systems.

  • With that noise behind the team, the focus returns to velocity, shelf productivity and revenue management.

  • Adjusted SG&A already fell to 26.4% of revenue from 33.6% a year earlier, and scale should keep pushing in that direction.

• Management has explicitly stated visibility back to low-50s gross margin on their last conference call, with locked aluminum conversion costs, the second North Carolina line ramping in H2 with full benefit in 2027, and price-pack architecture.

  • A ‘Q2 side-step’, then ‘Q3/Q4 stair-steps’ was also mentioned on the call, continuing into 2027.

  • In my opinion, this makes forward guidance the important piece in the next report, but Celsius does not usually release guidance, so a quick stock price dip after the next report wouldn’t surprise me if reported figures are lackluster.

So, overall, from 10% in 2026, the path to 15% by 2030 is a percentage point a year from gross margin normalization and SG&A leverage.

For reference, Celsius reported a 17% net margin in 2023 as a single brand, so I would imagine asking for a similar figure 5 years from now is not a crazy ask for a multi-brand organization.

This path produces net income to common of roughly $327M in 2026 rising to $764M by 2030, or EPS of approximately $1.27 in 2026, $1.77 in 2027, $2.15 in 2028, $2.54 in 2029, and $3.00 in 2030 on a ~256.5M share count.

• Note: total outstanding shares include 0.5% buybacks from 2027-2030, consistent with past shareholder buyback patterns.

Apply a 30-40x P/E — below Celsius’s own history, in line with Monster’s recent range — to $3.00 of 2030 EPS and you get $90-120 per share, versus around $30 today.

That’s a 28-36% CAGR, roughly 32% at the midpoint, or about 3.5x money over the period.

If the multiple only rerates to 25x, the 2030 price is still $75.

The sensitivities above show the full grid, including the downside corner, which even at 6% revenue growth and 8% net margins, the model produces around $46/share at 30x.

Bull vs. Base vs. Bear Cases

As a reminder, the forecast above was the base case.

Each scenario below is the same model with different answers to four questions: core-brand growth, Alani durability, international adoption, and cost/legal friction.

Bull Case:

Legacy Celsius reaccelerates into low-to-mid double digits as the reset space, Fizz-Free, and World Cup marketing convert and the Monster-after-2015 pattern plays out.

Alani keeps compounding on its awareness runway, international accelerates as Spain and Portugal replicate the UK’s penetration curve, aluminum pressure fades on schedule and Texas settles quietly like its predecessors.

My revenue path proves too low, margins reach the high end sooner, and 2030 EPS pushes through $3.50 with a growth multiple attached.

The P/S-based scenarios in the model ($159-179) live in this world, but I’m not underwriting them for my overall thesis. That is purely the cherry on top.

Base Case:

I’ve explained this case throughout the entire writeup, so I’ll keep this short.

Core brand in high single digit growth, Alani strong but maturing, international compounding steadily, gross margin stair-stepping to the low-50s through 2027, and lawsuits resolved for nuisance amounts.

This implies $90-120 by 2030, a 200-300% return.

Bear Case:

Core growth slips to 5% or lower and stays there, revealing the 6% as demand decay rather than integration noise.

International stalls as Europe regulates harder and the positioning doesn’t translate.

Texas metastasizes into a multi-state action that bleeds into Alani’s momentum, and value competition caps gross margin in the high-40s.

Celsius loses reset leverage with retailers as velocity flattens, and the platform stagnates as a roughly $4B revenue business at high-single-digit margins.

This translates into the $46-58 corner of the sensitivity grid, with a path toward the mid-$20s on the stock in 2026 before steadily moving upwards.

Even that scenario describes a profitable, cash-generative #3 platform inside Pepsi’s distribution system.

Not a zero.

What I’m Doing With My Money / Concluding Thoughts

Now the part I held back until the end!!!

I own 1,670 shares of CELH at an average cost around $31, which is currently a top-five position for me.

Management has been buying alongside me — in a single week in May, Fieldly bought 8,475 shares at $29.36, President & COO Eric Hanson added 7,500 at $29.04 (an 11% increase to his stake), and director Hal Kravitz added 8,400 at $29.73.

My plan from here is to trim part of my AMD position, my largest holding, into what I expect to be a strong stretch around its upcoming AI conference and earnings, and roll those proceeds into Celsius.

I could see this position growing another 50% from my current share count over the coming months.

It won’t reach the size of my AMD stake (nothing else ever will I don’t think), but it will sit firmly in my top three.

The main areas I’m watching as the time plays out is the core-brand growth reaccelerating in H2 2026, gross margins stair-step arriving in Q3 and Q4 as ‘guided’, Rockstar’s retail declines flattening as it completes its move into Pepsi’s system this summer, and international sell-through in Spain and the UK.

My takeaway is that Celsius has spent 18 months paying the full cost of becoming a multi-brand platform, with margin noise, shipment distortions, management distraction, and the market repricing the stock as though those costs were permanent and growth has permanently decelerated.

The KPI data we walked through says the consumer never left the brand. The shelf data says distribution is still expanding.

The model says the earnings power shows up over the next 24 months.

I believe the current valuation reflects temporary issues rather than permanent changes to the business, and if margins recover as management expects while growth stays anywhere near my base case, today’s price understates what this company earns by 2030.

I’ve followed this name for 7 years. This is the setup I’ve been waiting for.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

April and May were carried almost entirely by the AMD options, which did roughly 90% of the work in each month.

June was quieter and more balanced.

The portfolio returned +12.66% in June, added ~$170K, and finished the month at $1,514,688 in total account value.

The options still contributed most of the dollar gains, but the equity positions did more of the lifting than they have all year.

This month, I exited a position entirely and moved the proceeds into two other stocks, which I will walk through below.

One note before we start: this is the first monthly portfolio update going out to paid subscribers only. If you are reading this, thank you for being here. Founding member pricing is still available and locks in for life (20% discount).

Let’s get into it.

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

Read more

After reviewing earnings, industry data, valuations and recent developments, two companies continue to stand out above the rest in my portfolio for the remainder of this year.

The investment cases couldn’t be more different.

One is down nearly 40% YTD despite what I believe are improving underlying fundamentals. The market has focused on short-term headwinds, while I think it’s overlooking what the business could look like over the next 12-24 months.

The other has been one of the market’s best-performing stocks this year. Even after the run, I believe investors are still underestimating the magnitude of the opportunity ahead as new products launch, AI infrastructure spending accelerates, and several major revenue catalysts begin flowing through their financials.

In this write-up, I’ll walk through why I continue to own both positions, what I think the market is missing, the key risks to each thesis, and why I believe both still have meaningful upside from current levels.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor.

With a subscription, you will receive deeply researched stock deep dives, insight into my personal portfolio, along with each position I own and the related weighting, and much more.

Read more

Cash flow determines may things along the growth curve of a company’s lifecycle.

It dictates whether a company can pay their day-to-day bills, survive economics downturns, maintain purchasing power, take on debt, and even hire the right talent within the organization.

Most investors understand that cash flow matters, but far fewer understand how cash flow, reinvestment opportunities and leverage work together to create some of the greatest wealth-generating businesses in history.

Cash flow can be the sole reason certain investors will turn away from an opportunity.

Let’s get into it.

Subscribe now

Cash Flow Drives Equity Value

Free cash flow is the cash a business has left after covering its operating costs and the reinvestment needed to keep running — the cash actually available to the people who own it. The “take home pay”.

A share of stock is, at its core, a fractional ownership claim on the future cash flows of a business.

This is why finance theory dictates the intrinsic value of any asset is the sum of its future cash flows, discounted back to the present day (why people run DCF models).

You cannot pay dividends, buy back stock, fund M&A or pay off debt with net income. You can do all of those things with cash.

Earnings can be shaped in a dozen ways due to various accounting rules, but the cash a business actually generates is what ultimately sets the value of the equity, so for a growth company the central question is whether it produces consistent, growing cash flow.

Antonio Linares makes this point constantly:

Over long periods of time, stock prices tend to follow free cash flow per share

While earnings can fluctuate due to accounting rules and timing differences, it is ultimately the cash generation power of a business that determines equity value (equity value = market cap).

The Cash Conversion Cycle Creates the Growth Flywheel

If cash is what ultimately sets the value of the equity, then how quickly a business turns its sales into cash matters just as much as how much cash it produces.

The faster a company collects cash and puts it back to work, the faster it can grow.

Take the below for example:

Company B has a huge lag time before it can collect cold, hard cash from its customers.

Company A can immediately invest that cash it collects back into the business.

That is the most important point here.

Company A has a faster cash conversion cycle (converting inventory or a service into cash), which allows the business to reinvest back into the business more quickly.

This stronger cash conversion cycle creates a powerful growth flywheel.

Businesses that convert earnings into cash more quickly can reinvest that capital sooner into high-return opportunities — whether that’s hiring salespeople, expanding capacity, entering new markets, acquiring customers or funding strategic initiatives.

The faster cash is generated and recycled back into the business, the faster the company can grow. That growth produces even more cash, which can then be reinvested into additional value-creating projects, creating a compounding effect over time.

In many ways, the cash conversion cycle is a key enabler of scale.

Companies that can efficiently turn revenue into cash have a structural advantage because they can fund growth internally, reduce reliance on external capital, and accelerate the pace at which they compound value.

Why Software Gets Premium Valuations

This dynamic is also one reason why software and other asset-light businesses often command higher valuation multiples than industries such as retail, apparel, distribution or manufacturing.

Because these businesses require relatively little incremental capital to grow, they can reinvest cash more quickly and earn higher returns on invested capital (ROIC).

In contrast, more capital-intensive businesses often have cash tied up in inventory, receivables, equipment and working capital, slowing the pace at which they can reinvest and compound growth.

This is particularly true in software, where the cost of serving each additional customer is often minimal once the product has been built. As a result, successful software businesses can scale rapidly without a proportional increase in capital requirements.

—> This combination of efficient cash conversion, low incremental costs, and attractive reinvestment opportunities creates one of the most powerful compounding engines in business.

It is also a major reason why venture capital has historically gravitated toward technology companies and why ecosystems such as Silicon Valley emerged around them — these businesses can reach scale faster, generate ROIC and create significantly more value from each dollar invested.

While a strong cash conversion cycle and ROIC create the foundation for growth, debt can accelerate this process even further when used responsibly.

Amplifying the Growth Flywheel with Debt

Debt is simply a tool that allows a business to invest more capital into attractive opportunities than it could using internally generated cash alone.

I personally view debt as accelerated capital allocation — a way for a great business to put more money to work today than its own cash flow alone would allow.

—> If management can borrow capital at a cost below the return generated by the business, that spread accrues to equity holders, amplifying returns on equity (ROE) and accelerating growth. So, if the return exceeds the interest payments, it’s a profitable decision, all else equal.

In this sense, debt acts as a multiplier on an already attractive business model.

The growth flywheel sequence:

1. Start with high-ROIC business.

2. Generate cash.

3. Debt adds more capital.

4. That capital gets invested at the same high ROIC.

5. ROE rises.

The key caveat is that debt only creates value when the borrowed capital is reinvested into initiatives that generate incremental cash flow at returns above the cost of borrowing.

Simply adding leverage to a business does not improve the economics but rather increases risk.

In fact, under-utilized debt can become a drag on performance by consuming cash flow through interest payments while producing no corresponding return!

BUT, for a cash-generative businesses, leverage can create a powerful growth engine:

The business generates cash, reinvests that cash at attractive returns, supplements those investments with low-cost debt (additional capital), and then uses the resulting increase in cash flow to further expand the business and repay lenders.

—> This dynamic is one of the primary reasons us evil private equity people utilize leverage in acquisitions!

It accelerates the value creation generated by a business with strong underlying economics, and returns capital to investors more quickly (I promise we aren’t evil… only sometimes).

Here is something that I have noticed in the private markets that plays out similarly in the public markets:

Debt imposes a degree of discipline on management teams.

The obligation to service interest payments forces scrutinized capital allocation decisions, reducing the temptation to pursue ‘pet’ projects, empire building, or other uses of capital that fail to maximize shareholder value.

When used prudently, debt encourages management to focus on initiatives that generate the highest returns on invested capital.

BUT.

If cash flows deteriorate or reinvestment opportunities fail to generate the expected returns, you are still obligated to pay back interest payments + principal.

In these situations, leverage can quickly destroy equity value… and, in extreme cases, transfer control of the business from shareholders to lenders (also known as Chapter 11 / bankruptcy).

Thus, debt is most effective when paired with businesses that generate durable cash flows, strong unit economics and a proven ability to reinvest capital at attractive rates of return.

Debt does not create a great business, but simply allows a great business to become great faster.

The Ultimate Compounding

The best businesses share a common set of traits:

• Strong free cash flow generation

• Efficient cash conversion

• High ROIC

• Long reinvestment runways

• Prudent use of leverage

When these characteristics exist together, management can repeatedly turn every dollar of capital into multiple dollars of future value. Great for stocks!

Companies like Amazon are exceptional because they continuously convert capital into even more capital.

Ultimately, investors should focus less on reported earnings and more on how quickly a business turns earnings into cash, what returns it earns on that cash and how many opportunities it has to reinvest it.

Those factors are what drive sustainable growth and long-term value creation.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

AI is creating opportunity everywhere.

The AI infrastructure buildout, the physical robots about to walk into every warehouse, factory and eventually home on earth, autonomous land, air and maritime defense systems, the grid infrastructure rebuild, quantum computing, biotech and biology engineering breakthroughs, nuclear fusion, orbital and deep-space manufacturing.

This is a generational, once-in-a-lifetime setup that has the ability to create real wealth for the people actually paying attention. I've never been more excited to be an investor.

And I want to help you capitalize on it.

Subscribe now

I've gone back and forth on this post more times than I'd like to admit. The last thing I want is to sound like every other guy on here who slaps up a paywall and acts like he's doing you a favor.

Let me explain this in the best way I’m able.

For 8 months I've put everything out for free — the deep dives, earnings notes, monthly portfolio updates and others.

I stepped it up last month and went all the way and showed you:

• Every single dollar I've got in my portfolio

• Where it’s allocated

• How it moved from the prior month

• Various analysis/insights on portfolio performance

• How much is in my individual vs. Roth and why

This is sensitive information. But I want to commit to being as transparent as possible and allow this community to follow the portfolio for years to come.

And at the end of the day, my following will all come down to performance. And I’m confident I can beat the market. I’m already dominating the index the previous 3+ years…

I’ve built the institutional analysis skills through investment banking and private equity, and have paired these skills with almost 8 years of investing in the public markets.

But none of that matters to you unless it actually makes you money and makes you a better investor.

You don’t need an institutional background to beat the market anymore. Times have changed.

It’s always a great strategy to follow someone’s portfolio you trust, who lays out their work in public and who has consistently outperformed the index YoY.

—> That is what I am offering here on Substack.

And I’m playing with real money.

Here you can watch a portfolio grow (and shrink!) in real time — and see why every buy, sell and hold actually gets executed.

Here you can build your own instincts right alongside mine — so that a year from now you’re not just following me, you’re a sharper and more confident investor capitalizing on the biggest wealth-creation window any of us will ever see.

So with that — here is what changes going forward for Paid Members vs. Free Supporters:

At the end of the day, I want this Substack to be something I can run for years without burning out and dumping random AI takes in your feed just to keep posting.

One piece a week.

And a portfolio you can follow right alongside me as it appreciates in value — so over time I can actually help you get better at this and, yeah, help you beat the market.

Almost nobody on this platform opens their whole portfolio like this. I do.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

April was a fireworks show — the best month on record for the portfolio. The AMD leaps did roughly 90% of the work and the account crossed seven figures.

May answered it — and beat my expectations.

The portfolio grew +69.68% and finished the month north of $1.3M — the AMD leaps again carried roughly 90% of the portfolio gains. The leaps are now deep in the money and I am thinking about taking profits here over the summer. Still waiting for the San Francisco Advancing AI 2026 Summit.

Quick note before we get into it: this is the last monthly portfolio update going out to everyone. Starting next month, the full breakdown will be available only to paid supporters of ThePrivatePublicInvestor.

Founding member pricing locks in at a discounted price for life and is now live for a limited time. More at the end.

Let’s get into it.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

May Compounded at +69.68%

The portfolio returned +69.68% in May.

I added meaningfully to Celsius during the month and started exiting the last of my Micron, as the memory trade is louder than ever and I believe expectations are being steeply priced into the market. And when a downtrend begins in memory, it will be quickly and aggressively red.

The rest of the portfolio did not move much except one new add which I will discuss below.

Portfolio Holdings Update

I sold a piece of my December 2026 AMD calls this month and rolled the proceeds into shares. This is just me deliberately stepping down in leverage to be more conservative as the leaps are up +500-700% across the board.

“You don’t fold a winning hand, but you can slide a few chips back to your stack while it’s still your hand to play.”

What I’m not doing is selling the thesis.

I’m still holding 29 contracts across the 2026 and 2027 expirations, and the plan is to carry them into late July or later.

AMD’s Advancing AI 2026 Summit in San Francisco is exactly the kind of event where historic partnership announcements and a next-gen roadmap reveal make sense

It gives the company / stock some time to cool down after the massive run-up as well.

The only thing that pulls me out earlier is a genuinely silly move — think $700+, where the math forces me to close most of the options contracts.

And to be clear about where we actually are: even after this run, AMD’s GPU business is nowhere close to priced in.

The April move was data-center strength and the market waking up to a GPU-to-CPU ratio in AI buildouts approaching 1:1.

The bigger thesis, in conjunction now with the massive CPU shortage and identity shift, is the Helios-rack architecture rollout and the MI400 ramp in shipments.

None of these have pulled through the P&L just yet.

More partnership announcements and more transparency at the July event are the next catalysts.

Separately, the software space has been a graveyard this year.

I own 3 separate SaaS exposures in my portfolio, with one added in May.

I started a position in ServiceNow and believe it is one of the most undervalued SaaS companies in the entire market with a wide moat.

I trimmed a slice of Adobe to free up some temporary cash for opening a position in ServiceNow.

I intend to hold these 3 SaaS plays for years, and I’ll rank them here by the durability of their moat:

1. Veeva Systems

2. ServiceNow

3. Adobe

Veeva is the FDA-validated operating system for life sciences. ServiceNow is becoming the enterprise workflow engine. Adobe is the enterprise document and creative standard.

I’m spreading exposure across all three rather than betting on one. I know this may go against my core thesis around concentration, but I genuinely think all 3 of these names will do well for years to come.

Everything else is largely as it was. Coinbase, Nike, and the other core long-term holds — Estée Lauder, Amazon, Meta, BitMine, ASP Isotopes — are unchanged in size.

The thesis on each is intact and none moved enough this month to earn a fresh write-up.

Just a quick note that I’m looking at reallocating my Estée Lauder position toward a different, overlooked beauty company with strong growth potential. If you’re interested in the ‘why’ behind that trade, I’ll be covering it in depth for my paid subscribers.

A Record Year of Realized Gains: +$145K

YTD, the portfolio is up +174.54%.

Extremely grateful for such a tremendous year so far, and thank all of you for supporting me along the way!

But the number I want to flag here isn’t the headline return — it’s the realized one.

Realized gains hit $145,240 YTD, against $15,240 in losses.

That’s more than double the figure from the April update, BUT it’s mostly the product of deliberate rebalancing: taking profits on AMD options and rolling into stock.

I also sold more shares of Micron and have been focused on recycling that capital into positions with more road ahead of them.

2025 in Review

For context, 2025 closed at +51.78% with a 95% gain/loss ratio at +$94K.

A choppy three quarters were carried by a Q4 rally that peaked near +100% in mid-October before consolidating into year-end.

That 2025 chart is the reminder that drawdowns happen and consolidation occurs, but staying patient with your conviction pays off. And sometimes the market is very sudden.

And to clarify this point:

April 2026 was a moment, NOT a baseline.

2025 is more reflective of a normal year in the market.

My annual goal, on average over time, is two-fold.

• Beat the market and/or

• Achieve 15%+ annual return

If both happen, I’m a happy camper.

Where My Next Big Dollars are Going

Here’s what’s on my mind as AMD runs.

When I do trim that position in size, it throws off a large pile of cash, and that cash will need a home.

So I’ve been spending my research hours on the next thing, and I keep landing in the same two places: grid infrastructure and robotics.

Grid modernization is the physical hardware and the distribution-edge software that let aging utilities carry the massive AI compute load. Much of it serves the small and rural operators the giants overlook. The clearest expression of that on my watchlist is Tantalus (TGMPF).

Regarding the robotics sector, my preliminary Humanoid Market Mapping is already live, with a first look at the thesis below:

Below are other names I am keeping my eye on.

None of these are positions yet.

My watchlist will be continually updated for Paid Supporters.

Paid Subscription Detail

As mentioned in the April portfolio update, this is the last monthly portfolio update going out to all subscribers.

After that, the full breakdown moves to paid only — cost basis on every position, % gain per name, sizing as a percent of portfolio, real-time alerts when I add or trim, and where I am looking next to add to my portfolio.

Deep dives and other select posts have already gone paid.

Founding member pricing is still available, which locks you in for life as a discounted rate.

Subscribe now to keep your spot: https://theprivatepublicinvestor.substack.com/0b128dba

Macro Catalysts this Month in June

• Wednesday, June 10: May CPI data (the most critical inflation report of the month) and Wholesale Inventories

• Thursday, June 11: Producer Price Index (PPI), Initial Jobless Claims, and EIA Natural Gas Inventories

• Friday, June 12: University of Michigan Consumer Sentiment report

• Tuesday, June 16 - Wednesday, June 17: FOMC Interest Rate Decision & Press Conference

• Thursday, June 18: Continuing Claims and Initial Jobless Claims

• Tuesday, June 23: New Home Sales data for May

• Thursday, June 25: Final Q1 GDP Revision and Corporate Profits data

Prefer the polished PDF version? LINK HERE

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

If you enjoy ThePrivatePublicInvestor, it would mean the world to me if you invited friends to subscribe and read with us. If you refer friends, you will receive benefits that give you special access to ThePrivatePublicInvestor.

How to participate

1. Share ThePrivatePublicInvestor. When you use the referral link below, or the “Share” button on any post, you'll get credit for any new subscribers. Simply send the link in a text, email, or share it on social media with friends.

Refer a friend

2. Earn benefits. When more friends use your referral link to subscribe (free or paid), you’ll receive special benefits.

• Get a 1 month comp for 1 referrals

• Get a 3 month comp for 3 referrals

• Get a 6 month comp for 10 referrals

Visit the leaderboard

To learn more, check out Substack’s FAQ.

Thank you for helping get the word out about ThePrivatePublicInvestor!

This is the last free Deep Dive before paywalls are put up — only paid subscribers will have access going forward.

The United States is undergoing its fourth Industrial Revolution.

• First: Steam power and early factories.

• Second: Electricity, steel, and mass production.

• Third: Computers, microchips, and the internet.

• Fourth: Artificial intelligence and robotics

Artificial Intelligence is already in full swing.

Robotics is not.

AI deployed into the physical environment we live in is still in development, and I want to be early.

The advancement of LLMs and GenAI has catapulted progress in the robotics field by accelerating how machines learn, reason and react to the world around them through simulation and continuous incremental context.

The humanoid race is beginning and there is intense global competition among nations to develop these autonomous robots.

But first, why humanoids? Why do we need robots built like humans?

There sure are industrial cases where robotic arms and wheeled robots are needed in a manufacturing facility.

—> The answer is that our world is already designed for human bodies.

Jensen Huang said this best:

“Robots look like people because… the most important reason, is that we built the world for ourselves, and so the workstations of a factory, the manufacturing of a factory, was really created for people.”

Furthermore, because we share the same physical hardware, we have an unmatched dataset for training these machines to interact with the world, use human-centric tools, and operate the vast array of machinery built for our hands.

Walk the floor of almost any tech conference in 2026 and you’ll trip over a humanoid robot pouring coffee… an easy way to walk away convinced the future has already arrived.

Global humanoid installations came in around 16,000 units in 2025, up from essentially zero in 2023.

This is small, yes, but adoption is rapidly increasing.

KraneShares’ KOID, the first U.S.-listed humanoid ETF, launched just last June, and it has already been one of 2026’s standout thematic performers.

I think it’s time to dig into this space and find some standout multi-baggers.

Let’s jump in.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor.

With a subscription, you will receive deeply researched stock deep dives, insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling, decision-making investment guides, and the personal chat feature for the community.

Subscribe now to receive ‘founding member’ pricing and exclusive benefits before we rollout standard pricing for the rest of the community here on Substack.

With that, enjoy this piece and let me know if you have any questions!

Robotics is Going to Have a ChatGPT Moment

Jensen Huang has spent a year telling everyone that physical AI is the next multi-trillion-dollar wave and that robotics is about to have its ChatGPT moment.

What’s new in 2026 is that retail finally caught the scent — discords, comment sections and early green signals in public securities have seen notable increased activity and volume.

The humanoid robot story is not a "who builds the best robot" story. It is, at least for the next five to seven years, a supply chain story, and the supply chain has very specific chokepoints that specific companies own.

What Makes up a Humanoid Robot?

A humanoid is simply a battery, a brain, a body, and a bunch of motors, gears, screws and other parts connecting it all together.

Read more

It’s the longest I have consistently tracked a name in the market on record.

And I think the time is now to build out a position, especially at the currently discounted valuation.

This Q1, Celsius delivered a double beat, yet the stock sold off.

While the integration of Alani executed near flawlessly, that was overshadowed by the core Celsius brand losing market share to Alani Nu. Domestic dependency on Pepsi is also deepening and near-term margins are suddenly feeling tariff whiplash in aluminum.

The past couple of quarters have fundamentally made investors re-think their thesis and where future growth will be driven from. The core Celsius brand? Alani primarily? Will Rockstar ever get on its feet? Is international growth a primary growth strategy?

I’ll tell you what I think.

Let’s jump in.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor.

With a subscription to the ThePrivatePublicInvestor, you will receive deeply researched stock deep dives, insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling, decision-making investment guides, and the personal chat feature for the community.

Subscribe now to receive ‘founding member’ pricing and exclusive benefits before we rollout standard pricing for the rest of the community here on Substack.

With that, enjoy this piece and let me know if you have any questions!

Financials

• Revenue landed at $782.6M, +138% YoY

• Adj. EPS came in at $0.41 vs. the Street at $0.29-0.30

• Adj. EBITDA came in at $195.5M, beating consensus by ~28% and the corresponding margin expanded 3.8% to 25.0%, recovering from the Q4 trough at 18.6%

Per normal, no formal full-year 2026 revenue, margin or EPS guidance was issued or updated.

The stock closed +4.45% on earnings day before drifting back toward $30 over the following session(s). That reaction tells me the street is “iffy” on the future roadmap of Celsius, which is exactly where asymmetric trades are born.

Integration

Gross margins suffered this period, down to 48.3% from the low-mid-50’s in Q1-Q2 2025. The primary reasons for this GM compression was increased promotional activity, aluminum price increases and increased freight costs from integration commotion and weather-related drama.

These negative external market dynamics coupled with the Alani/Rockstar acquisition integration have impacted the overall margin profile this past quarter — there’s been a lot of short-term noise with Celsius’ business.

Regarding integration, the Alani Nu synergy target of $50M was captured in full, with management formally calling the integration complete on the call rather than pointing to it as a forward milestone.

In addition, distributor termination cleanup that was hanging over the balance sheet at $264M in December dropped to $40M by quarter-end, with $224M paid out in cash.

Great news!

Jarrod Langhans (CFO) also said the following on the call:

"We do have visibility into the low 50s and beyond with the initiatives and the programs that we have in place... probably as we look at 2026 in particular, Q2 is probably more of a side-step-type activity and then Q3 and Q4, where you're going to see the stair step and then continue on to 2027 with further stair steps."

So the back-to-50’s gross margin plan is structurally there, but the exact timing of this inflection will depend on external market factors — raw material costs (aluminum) and freight costs.

And get this.

Adj. EBITDA reached $195.5M, beating the Street's estimate by nearly 28%.

Gross margin was down during the period but Adj. EBITDA beat by 28% (and no, these weren’t caused by any crazy adjustments).

This means that operating leverage is already being realized post-integration on a cost basis, but revenue hasn’t caught up yet because the Company has been so focused on making the past 6 months of acquisition / integration as smooth as possible.

Now that Celsius has these brands mostly integrated (Rockstar still pending), the flow-through of the topline to the bottom line will inflate EPS meaningfully over the next 24 months.

The Multi-Brand Portfolio

The legacy Celsius brand grew +6% to $348M while Alani Nu came in at $368M, up ~60% PF.

On an LTM basis through Q1 2026, legacy Celsius and Alani Nu are roughly co-equal contributors at approximately $1.25-1.35B each. On a run-rate basis, Alani Nu has already passed legacy Celsius.

Rockstar landed at $66.6M with retail sales down 13% YoY, and Fieldly was clear that 2026 is a stabilization year on that brand rather than a growth year. The team has been focused on Rockstar SKU reconfiguration and resetting the brand maturely, rather than raw growth during integration.

And if you’re unsure about why Celsius acquired Rockstar in the first place, Rockstar essentially gives the portfolio access to convenience channel, gas station shelf and the male traditional energy consumer, which are areas where legacy Celsius and Alani both underpenetrate.

The combined market share of the brands of ~21% sounds great, but the legacy Celsius brand falling about 90 basis points YoY, from 10.8% to 9.9%. Not a great number to see. It’s what the business is built on.

But, the business model is changing, and Celsius is becoming a multi-brand portfolio.

The acquisition of Alani does show management is intelligent about competition and is focused on growing the business into a multi-brand portfolio long-term.

International Growth

International revenue came in at $35.3M, up 55% YoY, with Europe doing the heavy lifting at $25.4M (+36%) and the Asia-Pacific market a smaller base — $7.0M, and growing rapidly YoY at +212%.

That 55% growth looks fantastic, but some investors realized international share of total revenue shrank from 6.9% to 4.5%. This may seem troublesome, but this is simply because the U.S. business is growing that much faster, especially with Alani still on its rapid growth curve.

As of recently in March, Celsius formed an exclusive distribution agreement with Suntory Beverage & Food Spain, with Portugal next on the European footprint under the same partnership.

Fieldly framed the approach as one built on “key markets, strong local partnerships, disciplined launch plans and sustained marketing and distribution support.”

This is literally the SAME EXACT playbook Coke/Monster executed on — find a strong local distributor that already has the route, lean on their reps and skip the cost of building your own international infrastructure from scratch.

It’s a capital-light approach and ramps faster.

For now, I view the international growth plan as set-up to succeed — and just getting started.

What Am I Doing with My Money?

Celsius is currently a top 5 portfolio holding in my portfolio.

I plan for it to be a top 3 position sometime during the next few months, which means I will be continuing to accumulate shares.

This recent earnings report wasn’t great or bad. It was just good.

And good is where asymmetric opportunities are born.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

Yes. I will be the first to say that I was too early in Nike.

I was overly excited with the appointment of CEO Elliott Hill and the turnaround story of the legacy brand.

I knew it was a turnaround and I knew turnarounds take time, but I underestimated what a distribution-led turnaround actually requires — which is to say I underestimated almost everything that matters about how Nike runs its business.

When a company’s competitive advantage runs through how it sells, where it sells, who it sells to, and how that ecosystem reinforces itself over decades, the rebuild isn’t a turnaround in the usual sense.

It’s an overhaul of the operating model itself, and that takes years… not quarters.

Q3 FY26 (CY Q1) was the quarter that made that reality unavoidable for me, and it’s also the quarter that, in my read, marked the floor.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor.

With a subscription, you will receive deeply researched stock deep dives, insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling, decision-making investment guides, and the personal chat feature for the community.

Subscribe now to receive ‘founding member’ pricing and exclusive benefits before we rollout standard pricing for the rest of the community here on Substack.

With that, enjoy this piece and let me know if you have any questions!

The Beat Nobody Cared About

The earnings report itself wasn’t bad.

• Revenue of $11.28B came in ahead of $11.24B

• EPS of $0.35 beat the Street’s $0.29

• Gross margin landed 0.4% above consensus.

None of that mattered, because the Q4 guide was ugly.

Management estimated Q4 revenue (next fiscal quarter) to be down 2-4%, against Street estimates calling for positive growth. Greater China was also guided down approximately 20%.

JPMorgan downgraded the next day and cut its price target from $86 to $52, and twenty other analysts followed inside two trading sessions.

Not good.

The stock fell more than 15% the day after and hit a 12-year low of $43.17 forty-eight hours later.

12 years ago was when the iPhone 6 launched…

Also not good.

A distribution-led turnaround that's also a brand reset is structurally different from the kind most retail investors are used to seeing.

When the problem is the product, you can fix it in a design cycle. 18 months and a few good launches and the story can turn around.

When the problem is how you reach the consumer and what they think of you when you get there, you're untangling years of decisions compounded together over time.

Too much sold direct. Too little shelf space at wholesale. Aged Classics clogging the marketplace. Wholesale partners lost trust. Digital pricing that trained the consumer to wait for a discount. And a brand that drifted from sport toward lifestyle while HOKA and On built credibility in the lanes Nike used to own.

Every one of those issues feeds the others, and you can't fix one in isolation without making another worse.

• Pull sell-in down to clean the marketplace and revenue plummets.

• Push promotions to move inventory and you reinforce the discounting habit.

• Rebuild wholesale relationships and DTC margins suffer during the transition.

This is the spot Nike has been in, and the slowest-moving piece is Greater China, where Matt Friend (CFO) was honest enough this quarter to call the challenges "structural".

I think this statement rattled some investors, and honestly, Friend fback itand the team should have put that on the record a year ago.

Where is the Floor?

Underneath the ugly numbers and guidance, the operational picture IS doing what the team originally explained during the announcement of the Nike “Win Now” strategy.

North America gross margin fell (3.6%) on the surface but absorbed a (6.5%) tariff hit in the process, which means the underlying gross margin actually expanded by roughly +2.9% in the most important geography Nike operates in.

North America wholesale also grew 11%, Running was up over 20% globally, and February was the first month in 2 years where every channel in North America posted positive growth simultaneously.

Friend (CFO) gave the market the one number I wanted on the call: gross margin expansion begins in Q2 FY27, with Q1 FY27 as the final quarter of material tariff headwind.

As a reminder, Q2 FY27 for Nike is Oct-Dec of CY2026 (Q4)

I’d watch for Nike to deliver there.

In addition, late this year is when the “Win Now” actions are scheduled to be complete and the deliberate revenue drag from clearing aged, underperforming inventory falls off.

Elliott Hill (CEO) closed the call with a Camp Nou metaphor (FC Barcelona’s home stadium). Skim the quote if you don’t want to read in its entirety:

“Before we move to your questions, I want to leave you with an image that stayed with me from this quarter. I was in Barcelona meeting with athletes and leaders from FC Barcelona, a partner of ours since 1998, and I stood on the pitch at Camp Nou. If you’ve ever been there, you know it is more than a stadium. It is one of the most imposing stages in sport, a place built for pressure, belief, and unforgettable moments. And right now, Camp Nou tells another story, too.

Above the pitch, there is scaffolding. In the corners, there are cranes. Entire sections are unfinished. The stadium is being rebuilt tier by tier, piece by piece to accommodate over 100,000 supporters. The work is still underway. Capacity right now is limited and it requires patience and perseverance. And still, the supporters are in full voice. The players are still stepping on to the pitch focused on competing and winning. And all around them, the work is transforming what their home, their club will become.

What stayed with me was the reality of both things being true at once, competing today while building for tomorrow. FC Barcelona did not choose between performing in the present and preparing for the future. They are doing both at the same time. Standing there, looking up at the half-built stadium, it occurred to me, this is NIKE right now. We are taking deliberate actions that we believe will restore the health and quality of our business even when these actions create pressure in the near term.

We are removing what is not working. We are rebuilding parts of the foundation that needed to be rebuilt. And at the same time, we are continuing to innovate, to compete and to create for the future. That takes conviction. It takes patience, it takes belief, and it takes focus. Camp Nou is being rebuilt for the next — not being rebuilt for the next match, it is being rebuilt for the next era.

That is exactly how I think about the work we are doing at NIKE. I came back to help return this company to greatness and to build it the right way for the long term, to protect what has always made NIKE special and to modernize it for a new generation of athletes and consumers. So while the work is not finished, the direction is clear. Our focus is clear. And our comeback is within reach. And this fall, we look forward to sharing a fuller view of that path ahead at our Investor Day.”

That’s a fantastic, confident and patient response after a tough investor call.

Hard not to respect that response from Elliott Hill. He’s handling the pressure the right way.

That’s a CEO who isn’t pretending the timeline went as planned, and the recent insider buying could back it up.

• Hill bought roughly $1.0M of stock at $42.27 on April 13

• Tim Cook bought $1.06M at $42.43 three days earlier

There has been notable selling as well.

Bill Ackman (Pershing Square), who had been long on the turnaround since 2024, exited Pershing Square’s entire Nike position in late 2025 at approximately a 30% loss.

This is a contradiction worth naming. Trying to say only bulls are showing up at the bottom would be incorrect.

What I’m Actually Doing Here

Nike isn’t the asymmetric opportunity I thought I was buying a year ago.

It’s a staple consumer brand in the middle of a multi-year operational rebuild, and the stock is going to move slowly upward through late Q3 and Q4 of this calendar year as the “Win Now” actions complete, inventory health normalizes, and the China drag stabilizes.

I believe the floor on this business is meaningfully higher than the recent trading has implied.

Now, none of that adds up to a stock that doubles in twelve months.

Rather, it equates to a stock that grinds back over the next eighteen.

So I’m long, but I’m not adding, and I’m not selling.

Nike has become a value anchor in the portfolio rather than a high-conviction growth bet. It’s a position I hold because the floor is visible.

Sure, there are better short-term opportunities, and I’m putting capital toward them elsewhere. But every portfolio needs a couple value names to anchor the portfolio. Nike has become one of them for me.

The work is taking longer than I’d like, too.

But the direction is clear.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

The portfolio returned 92.98% in April and crossed $1M for the first time. AMD leaps did roughly 90% of the heavy lifting.

And yes — this really is a coincidence. I’d already planned to start monthly portfolio updates this spring, before April happened. The timing is genuinely coincidental and honestly a little inconvenient, because every future month will look pedestrian by comparison.

Quick note: Paid subscriptions are coming soon. The next monthly update will be the last one going out to free subscribers — more on what changes at the end.

One disclosure up front: I haven’t publicly disclosed my long-term AMD or Nike options before now. They aren’t core to my investing strategy and this was a one-off opportunity, and the most confident I’ve ever been in a position.

Too good to pass up.

The leaps have since exploded.

Let’s get into it.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With that, enjoy this piece and let me know if you have any questions!

April Compounded at +92.98%

The portfolio compounded 92.98% in 30 days.

The total account value of the portfolio almost reached $800K by month-end, with $45.5K in cash and $6K in crypto.

The inflection started April 23, where AMD exploded, inflating the leaps on AMD by near 100%. Micron also began its second wind, having generated a 100%+ in the past month. AMD leaps did the heavy lifting, contributing ~90% to the total portfolio gains during April.

Since my options have driven a majority of the gains in April, let’s look at how my portfolio would have done without the owning the AMD leaps:

In the image above, I am taking just my individual portfolio performance vs my Roth, since I house all of my options in my Roth IRA (seen in holdings screenshot blow).

Even stripping-out the options, the portfolio still returned ~29% vs. the market at ~10% during April.

The AMD Position I’m Not Selling

I’m holding 33 AMD call contracts across four expirations: $300 strikes for Dec 2026, and $350 strikes for Dec 2026, June 2027, and Dec 2027.

My blended cost basis approximately $35.08 per contract, and the blended price of the options today is $185.87.

Am I selling? I am thinking about selling a portion and replacing it with stock. The Dec-26 expiration leaps will be my first sleeve of options to cash out.

But.

My thesis hasn’t even started playing out yet on AMD.

The April runup came from data center strength and a shift in market psychology around the GPU-to-CPU ratio in AI buildouts approaching 1:1.

This is the warmup longer-term.

My actual thesis is the rollout of the MI400 and the Helios rack-scale platform. We haven’t seen a single dollar of topline contribution from this yet. No margin expansion. Nothing is flowing through the P&L, and I don’t think the analysts have been pricing in the actual demand this emboldens.

I keep asking myself: How am I supposed to sell when the thesis hasn’t delivered yet?

The hardware story is definitely already getting transparent, but the new tech hasn’t reached the customer yet.

If AMD gets this right and they continue closing the gap of ROCm vs. CUDA, they can easily compete with Nvidia on both axes, hardware AND software. That’s when this becomes a company that can rival Nvidia in scale, not just be the credible #2.

I’m holding all 33 contracts. I’ll flag the Chat if that changes and I decide to sell.

YTD Performance is +61.8%

YTD return is +61.8% through April 30, outperforming the general market by ~54%.

And YTD excluding the options contracts, the portfolio outperformed the market by 22%+.

• April YTD ThePrivatePublicInvestor Portfolio: +61.8% YTD

• April YTD S&P500: +7.9% YTD

Realized profit YTD sits at +$70,036 in gains versus $15,240 in losses (net +$55k) — I am continuing to rebalance the portfolio as big gainers such as Micron have exploded the past year, and I’ve decided to take gains on most of those shares.

2025 in Review

For context, 2025 closed at +51.78% with a 95% gain/loss ratio at +$94K.

A choppy three quarters were carried by a Q4 rally that peaked near +100% in mid-October before consolidating into year-end.

That 2025 chart is the reminder that drawdowns happen and consolidation occurs, but staying patient with your conviction pays off. And sometimes the market is very sudden.

And to clarify this point:

April 2026 was a moment, NOT a baseline.

2025 is more reflective of a normal year in the market.

My annual goal, on average over time, is two-fold.

• Beat the market and/or

• Achieve 15%+ annual return

If both happen, I’m a happy camper.

Current Portfolio Holdings

A few names worth flagging here:

1. Coinbase (49 shares): This thesis needs to be revisited with the CLARITY Act, but I am comfortable with the regulatory outcome and think Coinbase will succeed long-term.

2. Veeva (172 shares): I called the CRM attrition fear overblown two quarters ago. Holding this position for the long-term and will continue accumulating shares over the coming months. I would like this to be a top 3-4 holding.

3. Nike (363 shares + 20 leaps at $90 strike, Dec 2026): Nike has been a lagger, and these options have lost almost all of their value. Clothing is a sentiment-driven industry, and Nike is still viewed as a dinosaur. Hoping Nike posts a clean quarter before year-end that finally puts the turnaround on the tape. That’s when we exit the options. I will be holding the stock for the long-term.

4. Adobe (160 shares): Adobe has also been a lagger, but this red avalanche has occurred with the rest of the software names in the market. One thing to note: I am thinking of offloading some Adobe shares and loading up in NOW. Would like to have cash on the sidelines still and don’t want to be overexposed to software, as I already own Veeva Systems and Adobe.

5. Micron (15 shares): I’ve offloaded most of my shares in Micron this year, as I believe the sentiment on memory chip stocks is becoming overinflated and expectations will start to dramatically outweigh performance. At some point, any outperformance won’t be able to move the stock.

The rest of my positions in EL, BMNR, ASPI, AMZN and META have not changed in size and I remain bullish on these going forward into the remainder of 2026 and into 2027.

Paid Subscription Detail

The next monthly portfolio update is the last one going out to all subscribers.

After that, the full breakdown moves to paid only — cost basis on every position, % gain per name, sizing as a percent of portfolio, and real-time alerts when I add or trim.

Deep dives, earnings notes and market commentary go paid-first — with select pieces still released free.

Founding member pricing locks in for life and goes live before that transition (pricing detail to come).

Subscribe now to keep your spot.

Five Macro Catalysts in the Next Five Days

Wanted to note here at the end that this week is loaded with events, so stay tuned-in!

• Warsh confirmed as Fed chair

• April CPI (Tuesday)

• April PPI (Wednesday)

• April retail sales (Thursday) — same day Trump flies to Beijing

• Trump-Xi summit (Friday)

Prefer the polished PDF version? LINK HERE

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

AMD just printed an all-time high above $340 on Friday without a single piece of company specific news.

Intel blew out Q1 estimates and guided with a story of a severe shortage of CPUs — analysts immediately connected the dots and inferred that if CPU demand is melting up THIS hard, AMD is a HUGE structural winner.

AMD has certainly rallied dramatically from their semi-volatile, no-man’s land malaise since October of last year, but I believe this is only the beginning of AMD’s to-be, most dramatic rally in the lifetime of the company.

I am writing this article in real-time, as I am performing this analysis on the current valuation of AMD.

I didn’t prepare this analysis ahead — rather I am currently working through excel and conducting this analysis in real time.

I want to lay my thoughts out here regarding the current valuation of AMD, and where I realistically see this company’s stock price performing over the remainder of 2026 and forward.

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With a subscription to the ThePrivatePublicInvestor, you will receive insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling (coming soon), additional in-depth analysis on my portfolio holdings, decision-making investment guides, and the personal chat feature for the community.

As this channel grows, paid subscriptions will start to be incorporated, so subscribe early and stay subscribed to receive ‘founding member’ pricing and exclusive benefits.

With that, enjoy this piece and let me know if you have any questions!

The Street’s View on the Valuation

The majority of the market’s view on AMD is that it is considerably overvalued when compared with Nvidia, and I would highly disagree with this statement.

Let’s take a look at the below analysis I ran comparing AMD vs Nvidia on a P/E basis and P/S basis through 2028 based on street consensus.

Based on the current street view, AMD is overvalued on an earnings basis but trading at a discount on a revenue basis.

This makes sense given the current state of AMD’s business.

AMD has lagged Nvidia for the past several years, but has been picking up the pace in compute innovation and breadth.

With the rollout of the MI400 GPUs, the Helios rack-scale architecture, Venice EPYC and the Ryzen AI 400 series all hitting through the back half of 2026, AMD is on the precipice of the most aggressive product cycle in the company's history.

And this will be reflected in their revenue first and flow down to earnings afterwards.

The CPU shortage Intel just exposed will only add fuel to this rally. The Meta $60B / 6 GW deal, the OpenAI 6 GW commitment and Oracle's 50,000 MI450 Helios cluster ALONE imply a step-function in 2026/2027 financials the equity analysts are still scrambling to model.

Bank of America frames it as:

$15-20 billion of revenue per gigawatt of installed AI capacity.

With 12+ gigawatts already locked in across these three deals + implied incremental hyperscaler interest after the Intel earnings = the contractual revenue floor through 2028 is enormous for AMD.

I think we are at mile 6 of the marathon. Still 20 miles to go…

My 5-Scenario Model Build

I built a 5-scenario model — Low / Below Average / Base / Above Average / High — for FY2026, FY2027, and FY2028.

Read more

— Sir John Templeton

Let me be blunt.

Almost every bearish indicator I track is flashing red right now.

I always treat markets like this the same — I buy incrementally throughout the downturn and through the recovery.

Sometimes it may feel like you’re doing everything wrong in times like this.

That is normal for investors 1-10 years into their investing journey, including me (currently 7 years in).

I still feel like I’m doing things wrong sometimes!

But, for some reason, I picked up a good trait in the family and can stomach these downturns, however long and deep in the red (test me market!).

But I will say, I used to feel very worried in times like this.

But over time, incremental preparation compounding over time simmers those worries and gives comfortability in decisions.

I’m not here to scream doomsday, but I want to show you what the overall market dashboard actually looks like so you can make your own decisions.

Subscribe now

The Sentiment Picture is Ugly

The market is currently in a state of "Extreme Fear" (Index: 0-25), supported by deeply oversold technicals (RSI: 27) and a yield curve that has recently normalized to a +51 bps spread, which is historically a high-risk signal for an impending recession.

The VIX has been moving between 25-30 the last few weeks, remaining below the spike of 40 often seen at definitive bottoms.

I called this out earlier this week as a sign of the market potentially reaching a bottom in the coming couple of months.

The Indicators

The CNN Fear & Greed Index almost hit 10 this week, with the index sitting at 16 today, April 1st.

We are in deep Extreme Fear, the lowest reading of 2026, and comparable to the COVID crash and 2022 bear market lows.

Below depicts the LTM trend of the index.

Separately, the AAII Investor Sentiment Survey shows bearish readings at 49.8%, nearly 19 points above the historical average.

Bullish sentiment has sat below average for six consecutive weeks.

The University of Michigan Consumer Sentiment final March reading also came in at 53.3, which is a bottom 1st percentile historically.

People are scared.

Google Trends searches for “recession” and "stock market crash" have reached a one-year high, showing a significant acceleration as we’ve moved into Q1 2026.

Moody’s puts recession probability at 48.6%.

Goldman raised theirs to 30%.

Bear market talk is mainstream.

The CBOE SPX put/call ratio has also surged to ~1.3x, as it seems investors are aggressively hedging at the index level, which is the classic defensive posture you see during risky times.

The put/call ratio serves as a diagnostic for fear, measuring the volume of bets placed on a downturn against those betting on a recovery.

BUT, in the current climate, we are witnessing sentiment divergence:

• While the daily Volume Ratio has spiked as traders scramble for downside protection (buying puts), the Open Interest Ratio (total # of active contracts actually held in the vault) is hitting a 1-year low.

This tells the story of a “washout” in progress.

While the broader crowd is currently in a state of panic-buying insurance through purchasing puts, the smart money is quietly closing out their bearish positions and taking profits.

This could mean institutional big money is preparing for the clouds to break and the market to turnaround green.

I believe this means one thing, that has stood true historically:

—> Stay invested and prudently buy

The Technicals

The S&P 500 sits roughly 5-10% below its January all-time high (on the upper end if you discount the past couple days rally, and probably will go lower after Trump’s announcement after market close yesterday).

The index is trading below both its 50-day and 200-day moving averages, and a death cross confirmed around March 28-29.

We have since bounced from that death cross, which could be market support, but I don’t bet on moving averages. Algorithm traders (algos) could easily have been the reason for this short term bump the past couple of days.

The relief day rallies have been low-conviction, short-covering events that evaporate within a day.

More notably, the VIX closed at ~25 on April 1, up ~70% YTD from the 13-15 range where it started the year.

Historically, when the VIX spikes above 30-40, the S&P 500 has delivered strong forward 12-month returns

• Above 30% on average when VIX exceeds 40, with positive returns over 90% of the time since 1990.

We’re not quite at that 40 threshold yet, but we’re certainly knocking on the door.

The Macro Picture is Ugly Though

Oil first. The oil shock derived from the Iran conflict isn't just about gas prices.

Oil is used in plastics, fertilizers, pharmaceuticals, personal care products, clothing and more — it touches virtually every corner of the economy.

Those cost increases are already showing up on factory floors, with Chinese plastic suppliers raising prices roughly 15% according to CNBC.

And the full pass-through to consumer prices typically takes three to six months, meaning the worst of this inflation is potentially ahead of us, and is a reason why investors are pulling out and the market is red.

Core PCE already sits at 3.1%, which is well above the Fed's 2% target, while the labor market is softening (January added just 130K jobs; February shed 92K).

—> That puts the Fed in a brutal spot.

Futures markets have pushed the probability of a rate hike by year-end to 52%, but raising rates into a weakening economy risks tipping the U.S. market into recession.

The Fed cannot cut with inflation re-accelerating, and it cannot tighten without crushing growth.

This is CLASSIC stagflation — and the market hates it.

Stagflation: High unemployment + inflation + slow economic growth

Meanwhile, the yield curve has un-inverted after its historic 2+ year inversion, and this is actually a dangerous signal.

Recessions typically begin after the curve normalizes, not during the inversion.

We’re there now.

And then there’s AI.

The Magnificent 7 are down 20%+ from their highs.

Approximate as of 3/31/2026:

• $MSFT: Down 31-33%

• $META: Down 25-33%

• $TSLA: Down 25-27%

• $AMZN: Down 20-22%

• $NVDA: Down 19%

• $GOOGL : Down 15-20%

• $AAPL: Down 12%

The hyperscalers plan to spend $650-700B in 2026 capex on AI, yet we haven’t seen the margin pull-through yet.

The market is waiting to see this spend pay off with MASSIVE revenue acceleration and subsequently margin pull-through.

I personally think analysts don’t really understand the scaling of AI and what that actually looks like compounding over time (parabolic growth), but that’s a story for another time.

—> We are in a digestion phase of AI.

Note: Andrei Jikh has been my go-to economic “keep up-to-date” YouTuber I watch regularly, and I highly recommend everyone reading this to check out his channel.

https://www.youtube.com/@AndreiJikh

So What’s Priced-in?

A lot, actually.

All of what we just talked about is pretty common knowledge.

• Tariff uncertainty at current ~10.5% effective rates.

• The Fed being on hold (51% probability of zero cuts in 2026).

• AI multiple compression.

• SaaS disruption fear, possibly even overdone

• Consumer sentiment is already terrible.

What’s not fully priced in is a prolonged Strait of Hormuz closure pushing oil above $130 or higher, AND whether AI capex actually translates to revenue.

Those are the 2 big whammies the market is waiting on.

Now, I know what some of you are thinking.

This time is different. AI is transforming everything. The geopolitical landscape has fundamentally shifted. Fiscal policy dynamics are unprecedented.

And those folks are right!

Today’s market does have unique characteristics.

But people say “this time is different” every single cycle.

They said it in 1999 about the internet. They said it in 2007 about housing. They said it in 2021 about stimulus. They said it in 2022 with the inflation crisis. They said it during the tariff crises 12 months ago last April.

Human behavior does not change.

But, I’m not waiting for confirmation, as the market will have already moved dramatically up by then.

And if it’s bad news?

Great, I have cash to buy more.

Just remember: Stay invested, prudently buy into the market periodically, stay knowledgeable, be patient and stick to your plan.

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

I’ve realized THIS is what all the buzz is regarding “I now have a junior analyst and operations assistant living on my desktop!”

Cowork is where the real acceleration of workflow starts, and where we can enable tools to work for us in the background.

I usually wouldn’t post about something like this, but I feel it’s that important, and can save everyone reading this, hours during the work week.

In this post I’ll show you what Cowork actually is, how skills and plugins fit in, and a few concrete workflows you can steal for your own investing process and everyday work.

Subscribe now

What is Claude Cowork?

First let’s start out with what Claude Cowork is.

Claude Cowork is a platform capability that allows users to sync their drive and company information into one prompted conversation.

It’s essentially a mode inside Claude where the AI can plan and execute multi‑step tasks on your behalf instead of just responding one message at a time.

Anthropic describes Cowork as using the same agentic architecture as Claude Code, but packaged for non‑technical knowledge work right inside the desktop app.

Cowork works directly with a folder or project on your computer, reading, editing and creating files. This allows AI to continually live within your workflow, instead of starting from almost ground-zero within a chatbot.

I’ve been using this at work for a multitude of tasks such as data cuts, modeling questions, market landscaping, industry deep dives, company specific questions and other simple tasks like notes summaries, quick data pulls and general questions that require drive-level context (aka specific folders).

Doing these things regularly within Cowork has legitimately saved me 15-20 hours a week… and has provided more value at the same time.

For Substack, it's a great resource because the AI can continually learn from current posts, portfolio holdings, and general activity across the app.

It’s also great for measuring audience demographics and content-specific interests, and subsequently giving you trend-level feedback based on your highest or lowest performing pieces.

Regarding personal investing (the audience likely reading this), there are a multitude of ways you can leverage this, which I will get into later in this writeup.

I'll be the first to admit this is ironic given the writeup you're reading, but the best way to learn Cowork is to just buy it and start tinkering.

Doing always beats analyzing through instructions and tutorials, in my opinion.

Share

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

Application Plugins

Two capabilities that contribute to the efficiency of Cowork are skills and plugins.

Read more

— Shantanu Narayen

Contents

1. CEO Transition

2. FireFly & AI Monetization

3. Financial Performance

4. Innovator’s Dilemma

5. Watch Items

6. Bull Case

Subscribe now

CEO Departure

Let’s get this out of the way.

Yes, the CEO of Adobe has left.

The man who took over ‘Adobe Systems’ back in 2007 before the Financial Crisis.

The man who took Adobe through the financial crisis and built one of the most successful enterprise SaaS companies in the world.

Under Narayen's 18 years, Adobe grew from a sub-$1B packaged software company to a $25B+ cloud subscription juggernaut, multiplying the stock price more than sixfold versus ~350% for the S&P 500.

Narayen’s tenure is one of the greatest CEO runs in enterprise software history.

This does sting for long-term Adobe holders!

But, note that Narayen will stay on as Chair of the Board while the company finds his replacement.

I don’t think this departure is necessarily a bad thing. I think it’s natural, and will be a planned, orderly transition.

Narayen is approaching his mid-60’s, and with the world changing so rapidly today, I think it does make sense that he is resigning as CEO.

The board has appointed Lead Director Frank Calderoni to chair a special committee that will evaluate both internal and external candidates…. and the internal bench is deep.

The frontrunner in my view is David Wadhwani, President of Creativity & Productivity. Wadhwani was previously CEO of AppDynamics and actually helped architect the original Creative Cloud shift during his first stint at Adobe.

CNBC has previously reported that Wadhwani was being groomed for the role and former Adobe executive Brad Rencher called him “made in Shantanu’s image”

Wadhwani is currently the executive leading Adobe’s AI transformation and Firefly strategy, which is exactly the skillset needed for this next decade!

Anil Chakravarthy, President of Customer Experience Orchestration, is the other internal contender. He joined from Informatica and oversees the enterprise CXO business (smaller business segment).

Regardless of who gets the appointment, I actually think this is a net positive on a 2-3 year horizon.

Adobe needs a forward-thinking leader who will aggressively push AI-first product innovation and isn’t anchored to legacy business decisions.

Narayen framed the opportunity perfectly on the call:

“Our mission to empower everyone to create represents an even larger opportunity in the AI era.”

Narayen also highlighted that AI-first offering ARR more than tripled YoY, and that Adobe surpassed 850M MAU (monthly active users) across Acrobat, Creative Cloud, Express and Firefly (17% YoY growth).

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With a subscription to the ThePrivatePublicInvestor, you will receive insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling (coming soon), additional in-depth analysis on my portfolio holdings, decision-making investment guides, and the personal chat feature for the community.

As this channel grows, paid subscriptions will start to be incorporated, so subscribe early and stay subscribed to receive ‘founding member’ pricing and exclusive benefits.

With that, enjoy this piece and let me know if you have any questions!

Firefly Growth Ramping

Let’s talk about the AI monetization story because this is where the thesis gets hot.

Firefly ARR was $250M+ in Q1, with subscription/credit pack ARR growing 75% QoQ. Generative credit consumption also grew 45%+ QoQ, with video-generative actions surging 8x YoY.

Creative freemium MAU crossed 80M, up 50% YoY, which is a direct feeder in Adobe’s paid conversion funnel.

This is an intentional strategy for conversion.

The monetization architecture/strategy is smart.

First, users get a base amount of generative credits included in their subscription.

Once they exceed that limit, they buy credit packs or upgrade plans.

—> This drives higher ARPU without requiring new seat additions, which is exactly the model I highlighted in my original deep dive.

Even if just 15-20% of Creative Cloud users exceed base AI usage and spend an incremental $10-15/month, that’s hundreds of millions in high-margin revenue!

And remember, Adobe isn’t directly competing with the new, innovative AI models to realize this growth within their existing customer base.

Adobe’s partnerships with Runway (video AI), OpenAI (Sora integration in Firefly Studio), and Google (Gemini models) create a model-agnostic platform, so, if someone builds a better model, it still runs inside Adobe’s software.

Consequently, Firefly Enterprise new customer acquisition grew 50% YoY, with over 2,500 custom models created via Firefly Foundry since launch.

GenStudio and AEP & Apps ARR each ALSO grew over 30% YoY, and there are 650 customer trials underway for Adobe’s agentic web products (LLM Optimizer, Sites Optimizer, Brand Concierge).

The enterprise AI pipeline is building, and customers are buying in….

Q1 Performance: Beats across the Board

Now let’s dig into the general report.

• Total Adobe ARR: $26B+, +11% YoY

• Q1 Revenue: $6.4B vs. $6.3B expected — BEAT (+12% YoY reported)

• Q1 Non-GAAP EPS: $6.06 vs. $5.86 expected — BEAT (+19% YoY)

• RPO: $22.2B+, +13% YoY, which is strong forward visibility!

• Operating Cash Flow: ~$3B (+19% YoY), which is a Q1 record apparently!

• Non-GAAP Operating Margin: 47%+

That’s a great earnings report if you ask me.

Note: there was some tax noise I noticed in the EPS calcs as well, so stripping those out would have realized even more margin expansion and, thus, a larger EPS beat.

One thing I’d like to note here is the accelerating growth of Adobe, which is highly attractive for a mature enterprise SaaS business like this.

The below table shows that Adobe’s product suite is not becoming less attractive or archaic for business or creative users, but rather augmented from AI and AI model integration.

In more detail:

• Subscription revenue across both business segments hit ~$6.2B (+13% YoY)

• The Business Professionals & Consumers segment grew to ~$1.8B (+16% YoY)

• Creative & Marketing Professionals grew to ~$4.4B (+12% YoY)

• Acrobat AI Assistant ARR tripled YoY

• Express is now used in 99% of U.S. Fortune 500 companies

Guidance also came in pretty good with revenue of $6.4-6.5B (slightly above Street), non-GAAP EPS of $5.80-5.85 (above consensus of $5.68-5.70).

The Innovator’s Dilemma: Is Adobe Falling into the Trap?

Clayton Christensen’s Innovator’s Dilemma warns that incumbent companies get disrupted when they over-serve high-end customers while ignoring cheaper, good-enough alternatives that eat them from below.

Is Adobe falling into this trap? I’d argue no.

Adobe is doing something most incumbents fail to do, which is investing aggressively in the disruptive technology itself rather than protecting legacy revenue streams. They are also focused on democratizing access to this technology through their model-agnostic strategy.

Firefly, Adobe’e new AI platform for creative workflow, isn’t a defensive product, but rather an all-in-one creative AI studio with a broad portfolio of the industry-leading models, a $250M+ ARR business growing 75% QoQ and a clear future monetization framework.

Also, Adobe markets to everyone. Not just their high-value customers.

The one area where the Innovator’s Dilemma does have relevance in conversation is the Stock business decline.

Management admitted this erosion is “happening faster than planned”.

The traditional stock photography/content licensing business (about $450M ARR) is getting cannibalized by Adobe’s own Firefly tools, free AI alternatives and likely competitors.

This is a self-cannibalization trade-off, but I’d rather have Adobe cannibalize its own revenue than let Midjourney or Figma do it for them.

In case you missed it, here are some of the best investing insights for 2026:

• 5 Stocks Worth Your Attention in 2026

• Adobe & the AI Pivot

• Deep Dive on AMD

What to Watch Going Forward

Despite the strong numbers this quarter, there are legitimate risks that I am not agnostic to:

CEO Succession Uncertainty: Until a successor is named, there’s lost strategic at a very critical time.

An internal pick like Wadhwani would signal continuity and aggressive AI execution, which I think is the right move, but if this gets prolonged, it could extend uncertainty and a continued decline on the stock price.

Competitor AI Momentum: Figma, MidJourney and Canva are strong competitors in the space and are not slowing down.

Figma, for example, is growing at 40% topline and announcing strategic partnerships, such as with Anthropic, integrating Claude Code for AI-powered workflows.

MidJourney also continues to release newer and newer AI models on its platform, challenging industry incumbents like Adobe to keep up.

Canva, Adobe’s biggest competitor, hit $4B in ARR, growing 35% YoY, with 95% Fortune 500 penetration. Their anticipated H2 2026 IPO will bring additional capital and competitive visibility, encroaching on Adobe’s territory.

The SaaSpocalypse: This is the major reason why Adobe’s stock has taken a gigantic hit the past year. About $300B in software market value was wiped in early February 2026 alone!

$IGV, the iShares SaaS ETF, is down (~25%) the last 6 months.

The entire SaaS sector is getting re-rated to lower multiples based on fears that agentic AI will destroy seat-based licensing models, and anyone will be able to vibe-code their way through custom workflows.

I believe this risk is overblown, but it is yet to be seen with certainty whether these re-ratings are appropriate or justified.

Stock Business Erosion: As mentioned, this headwind is accelerating faster than planned and creates a drag on total ARR growth. But, Firefly is accelerating faster than the decline in this business and could successfully replace the traditional creative software offering going forward.

Cash Position & Buybacks: Adobe repurchased roughly 8.1M shares in Q1, leaving ~$3.9B remaining on the $25B authorization. The buyback program was approved in March of 2024, so they’ve burned through almost ALL of that in 2 years.

Adobe’s $25B buyback program will be exhausted by mid-2026, ~2 years ahead of its formal expiration.

Given 7 consecutive buyback program renewals, I would guess they will reinstate this program.

The Semrush acquisition does create a temporary funding pinch, but I don’t think this will be a huge deal considering the operating FCF that Adobe produces.

I’d be on the lookout regarding the size of this buyback program though, as historically, each sequential buyback program has doubled compared to the last, which would infer a $30-40B buyback program this go-around.

A more conservative sizing could acknowledge a tighter balance sheet and growing R&D dollars needed for AI investment.

I’m guessing this announcement will be during the June 2026 earnings call.

I’m Bullish and Staying Long

Despite the SaaS’pocalypse noise, I’m supportive of the bullish thesis on Adobe and think the following are justifiable reasons to invest / stay invested:

Commercial Safety Moat: Firefly is trained exclusively on licensed content with full IP indemnification.

Enterprises will need legal certainty for AI-generated content, and this is a massive differentiator versus Midjourney or other open source tools that can’t offer the same protections.

AI Growth Acceleration: AI-first ARR tripled YoY.

Firefly ARR is growing 75% QoQ, generative credits are growing 45% QoQ and MAU are growing 17% YoY.

The AI business is growing rapidly and faster than anticipated.

Cheap Valuation: Adobe trades at roughly 15-17x trailing P/E, which is about 63% below its 10-year average.

The forward P/E is lower around 11-14x and the PEG ratio sits at 0.90 (which below 1.0 is traditionally considered undervalued).

For comparison, ServiceNow trades at 53-73x trailing, Intuit at 28-30x and Salesforce at 21-25x.

Adobe is the cheapest large-cap SaaS stock by trailing P/E, despite generating $10B+ in annual operating cash flow and carrying 47% operating margins and an expanding bottom

Wall Street Doesn’t Like It

The analyst community is already cautious.

• TD Cowen cut its price target to $325 from $400 CNBC (Hold)

• Benzinga HSBC lowered to $302 from $388

• Goldman Sachs holds sell rating at $290

• Piper Sandler has warned that Adobe sits in software categories “among the least insulated from AI disruption”

• UBS reduced to $340 from $375 (Neutral)

• Stock Observer Citi trimmed to $315 from $387, expecting an uneventful quarter

• Oppenheimer had already downgraded to Perform in January

RBC Capital was among the few bulls, maintaining Outperform with a $430 target.

I like that the Street doesn’t like Adobe now… it only creates opportunity for more upside if the big money decides to jump in later!

Overall

Adobe is accelerating topline growth at 12% for the most recent quarter, with compounding EPS at 19% and trading at a valuation usually reserved for ex-growth businesses.

The math REALLY doesn’t add up unless you believe AI permanently impairs Adobe’s business.

I think these Q1 results suggest the opposite…

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

— Philip Fisher

The P/E Multiple: A Good Refresher or Helpful Guide

Most investors learn about the P/E ratio (“earnings multiple” or “multiple” or “P/E” for purposes of this discussion) in the first 5 minutes of their investing journey.

Price divided by earnings… seems simple enough!

But I’ve noticed over the years that most investors, usually the new’er’ retail investors, are still using it incorrectly years later… or at least looking at it from a narrow scope.

Today I want to fix that.

Let’s start from first principles, then discuss how to use the metric on a relative basis and dig into some one-off, use case strategies to make this metric more dynamic.

Feel free to skip around this post —> I know many of you are familiar with some of these concepts but maybe are newer to other concepts.

Let’s dive in.

Contents

1. What You’re Actually Looking At

2. The Forward P/E

3. Run-Rate Analysis

4. Growth Adjusting with PEG

6. Relative Expansion & Contraction

7. Comparing to Industry & Peers

8. When the Multiple Breaks Down

9. The Balancing Act of Putting it All Together

Subscribe now

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With a subscription to the ThePrivatePublicInvestor, you will receive insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling (coming soon), additional in-depth analysis on my portfolio holdings, decision-making investment guides, and the personal chat feature for the community.

As this channel grows, paid subscriptions will start to be incorporated, so subscribe early and stay subscribed to receive ‘founding member’ pricing and exclusive benefits.

With that, enjoy this piece and let me know if you have any questions!

What You’re Actually Looking At

When you pull up a stock on Yahoo Finance or Google and see a P/E ratio, you’re almost always looking at a trailing twelve months (TTM) figure.

That’s the current stock price divided by the last four quarters of reported earnings per share (EPS).

It’s backward-looking. And that’s fine as a starting point!

But the direction of the earnings multiple matters just as much as the ratio itself.

A stock trading at 30x earnings that was at 40x six months ago is telling you a different story than one at 30x that was at 20x.

• Is the multiple compressing because earnings are growing into the valuation?

• As in, the denominator is growing faster than the numerator

• Is it expanding because the market is getting ahead of itself?

• As in, stocks are trending to the upside, pushing up valuations

Tracking the trend helps tell the story of how the market has viewed the valuation of the business over time.

Pairing this ‘valuation over time’ with other aspects of your investment thesis such as your financial model (projections, KPIs, etc.) can help craft a more fulsome investment thesis.

So, remember to track the trend, not just the snapshot.

Forward P/E: The Market’s Expectations

Stocks are priced based on future earnings, not historical earnings.

The trailing P/E tells you what the market paid for yesterday’s business while the forward P/E tells you what the market is paying for what this business is expected to look like in the future.

—> This is arguably the most important variation of the metric in my opinion.

The forward P/E uses analyst consensus estimates for the next twelve months of earnings as the denominator (aka Wall Street’s estimates), rather than the company’s reported, actual last twelve months of financial performance.

This is why it matters:

Imagine a company just went through a rough year. Maybe they had company or a complete product portfolio restructuring to meet market demand.

The trailing P/E might look absurdly high, say 60x, because the company just had a bunch of expenses to realize on their P&L.

But, if analysts expect earnings next year to double because they are betting on or “pricing in” a successful turnaround of the business, the forward P/E could be 25-30x.

Those are two completely different investment stories. Ignoring the forward P/E because the trailing number looks “expensive” could mean you miss a turnaround thesis playing out.

But it’s not just turnarounds, that was just an example. Let me give you a better example:

Let’s think of a high-growth tech firm that just finished some heavy R&D spending on a new product. Its trailing P/E might be 100x because past earnings were suppressed by all of that R&D spending.

However, if the business is rolling out or just launched that new product, next year’s earnings could quintuple.

Let’s say the forward P/E being priced in is 20x: the stock is actually cheap relative to its imminent scale.

—> Using only trailing data is like driving a car by only looking in the rearview mirror; it tells you where you’ve been, but the forward P/E shows you the turn ahead.

Here’s my challenge to you:

Don’t just look up the forward P/E. Build your own.

Even a simple model works. Take your projected revenue, apply the margin you think is realistic, back into EPS, and divide the current price by that number.

Now you have your own forward P/E.

Compare it to the Street’s consensus. If the Street says 25x forward and your model says 18x, try to understand that gap.

• Are you more bullish on margins?

• Are analysts underestimating a revenue headwind?

• Are they being too conservative on a new product rollout?

Try this for practice even if it seems like a waste of time. Writing this down will help you conceptualize this concept.

This is the difference between reading someone else’s opinion and forming your own.

Run-Rate Analysis: An Alternative Lens

A strategy I’ve taken from my experience in the private markets and applied to valuing public companies is applying a run-rate to a company’s last fiscal quarter or quarters. This involves:

Taking the most recent quarter’s EPS, annualizing it (multiply by four), and using that as your denominator in the P/E ratio. This gives you the “last quarter run rate” P/E.

Why does this matter?

Because businesses change.

If a company just posted a breakout quarter, maybe margins expanded? Maybe a new revenue stream kicked in?

Personally, I like to use 2 quarters for businesses because it strips out possible single-quarter noise such as macroeconomic supply/demand changes or maybe something like raw material price volatility. Using the last two quarters smooths that out while still giving you something more current than the full TTM figure.

The TTM number is dragging along three older quarters that no longer reflect the current business.

The run rate P/E can give you a more insightful snapshot of what the business actually looks like right now.

Quick example: say a company earned $1.00 EPS over the TTM, but the most recent two quarters came in at $0.35 and $0.40. Annualizing those two quarters gives you $1.50 ($0.75 × 2).

• On a $40 stock, the TTM P/E is 40x, but the run-rate P/E is ~27x. Very different valuation!

You can also get creative here by stripping out one-time charges, normalizing for seasonality if the business is cyclical or adjust for SBC if it’s distorting GAAP earnings.

The point here is to get closer to the true earnings power of the business.

Try this as well:

—> Compare your run-rate P/E to the forward P/E

If the run rate P/E is 27x and the consensus forward P/E is 22x, the market is expecting earnings acceleration.

• Is that realistic?

• What needs to happen for the business to get there?

• Where is that margin expansion coming from?

  • Is there new technology in place that will help accelerate this?

  • Did they make recent hires in the previous year that will ramp during the next year(s)?

Ask these questions first.

If you think the margin acceleration is coming, you might have a compelling entry point.

If you think the Street is too optimistic, maybe wait it out. I’ve found this to be a useful quick-check in my analysis.

Note that run-rate analysis is only helpful for companies without dramatic seasonality. For example, this wouldn’t be super helpful for a company like Abercrombie if you are run-rating the last 2 quarters of the calendar year (inflated financials from the holiday season).

Run-rate analysis would be helpful for valuing a company like Adobe with more recurring-based revenue without seasonality.

Performing this run-rate analysis for metrics like ARR for tech businesses are a popular use case — but this run-rate analysis can be applied to multiples such as P/E as well.

One more practical note:

• If a company is days away from their earnings report, be cautious about making valuation judgments based on the current P/E. That denominator is prone to change short-term, and as investors, we don’t want to hinge on short outcomes.

In case you missed it, here are some of the best investing insights for 2026:

• 5 Stocks Worth Your Attention in 2026

• Adobe & the AI Pivot

• Deep Dive on Coinbase

Growth Adjusting Your P/E

Once you’re comfortable with forward P/E, the next step to rounding out our understanding of earnings multiples is asking:

“Is this P/E justified given how fast the company is growing?”

I personally use this ratio less than most people, but the PEG ratio, in theory, addresses this question on a relative basis.

PEG = P/E ratio ÷ expected earnings growth rate (expressed as a #, not a %)

So, broadly, a PEG below 1.0 suggests a stock may be undervalued relative to its growth, while above 1.0 signals the market may be paying a premium.

So, a stock trading at 30x earnings with 30% expected growth will have a PEG of 1.0

Let’s say you’re comparing two companies in the same industry providing a very similar product/service offering.

• Company A trades at a 20x P/E with projected 10% growth for the next year (PEG of 2.0)

• Company B trades at 35x with 40% growth (PEG of 0.88)

• On a P/E basis, Company A looks cheaper

• On a PEG basis, Company B is the better deal

• Company B looks cheaper because you’re paying less per unit of growth

This ratio is ONLY useful for comparing businesses in the same sector though, as a PEG ratio for a utility business would be much different than a AI-enabled SaaS business.

It does have its drawbacks due to it’s:

• Reliance on forecasts, which are not always accurate

• Ignores capital structure (cash and debt levels)

• Meaningless measure for smaller companies with little profitability

• Ignores businesses with cyclicality such as in energy or materials

The PEG ratio should be used as an additional metric to round-out an analysis, not a standalone measure of growth relative to earnings power.

Relative Expansion & Contraction:

I’ll start by saying I use this methodology a lot, as it’s great for relative analysis.

A stock can grow earnings 15% a year for five years, but if the P/E compresses from 30x to 15x over that same period, due to investors becoming bearish on the stock or future of that industry (like SaaS businesses in the market right now) your return is roughly flat.

Also, a company with “okay” earnings growth can deliver huge returns if the market re-rates the business, which can expand the multiple (this would be the inverse of the above, as investors would become bullish on the company or industry).

• An example of this is the “picks and shovels” of the AI infrastructure industry, such as businesses who provide cleaning services to fabs, or electrician businesses that service datacenters.

  • These businesses may not have had crazy historical growth, but the growth expectations are sky-high due to the future buildout needs of AI infrastructure.

Think about what actually drives multiple expansion:

• Improving growth expectations

• Business model transformation (transaction-based to subscription revenue, for example)

• Market sentiment turning positive on a sector

Multiple contraction is the opposite:

• Slowing growth

• Deteriorating sentiment

• Competitive threats

• Commoditized products

Understanding what could expand or contract the multiple is just as important as analyzing the earnings themselves.

This is also when comparing a company’s current P/E to its own historical range becomes very useful!

Let’s look at Adobe, as I think this company is undervalued currently, compared to where it’s traded historically on an earnings multiple.

Adobe is trading at a ~15x multiple right now, when it’s traded at a ~45x average multiple over the last decade! Even if we take a more recent 5-year average, the multiple stands at 40x!

This is where it comes down to your own personal conviction.

Some folks believe this business’s service offering will become commoditized, which will re-rate the SaaS multiple down, which has driven the stock price meaningfully down over the past ~6 months, hence the “SaaS’pocalypse”.

I don’t think this is true, which would lead me to believe that this P/E is offering up a great deal!

—> This is where real risk/reward can become asymmetric

Your total return from a stock is a combination of earnings growth and the change in the multiple the market assigns to those earnings.

Comparing to the Industry

Stop comparing against the S&P 500’s overall P/E as your benchmark.

The spread between sectors is massive.

Energy stocks might trade at 7-8x earnings while tech trades at 30x+!

Comparing your Sandisk semiconductor play to P&G’s P/E is comparing apples to oranges. I hope you get this one, haha. Always compare within the sector, and ideally within the specific sub-vertical and provide similar services (sometimes you will have to broaden your comparisons, but you get my drift).

Even within the same industry, you need to ask why a company trades at a premium or discount to its peers. A company growing revenue at 25% deserves a higher multiple than another business growing topline at 5%.

Margin profile matters as well. Same with capital intensity.

A software company trading at 35x with 80% gross margins is fundamentally different than a manufacturer at 35x with 30% margins.

Simply ask the question: is this premium or discount justified by the underlying business quality?

A stock trading at 2x its sector average P/E isn’t overvalued if it’s growing 3x as fast, and a stock trading at 20% of its sector average might be cheap for a reason (classic value trap)!

Dig into the why before making a judgment.

When P/E Breaks Down

Firstly, don’t use the P/E ratio when a company isn’t profitable or is barely profitable. It’s useless. A company earning $0.01 per share trading at $50 technically has a P/E of 5,000x. It doesn’t mean crap.

Think of Tesla back during their growth cycle in 2020 when their stock skyrocketed 700% —> they just became profitable that year and were valued at $750B+ by the end of the year!

For early-stage or high-growth companies that are reinvesting aggressively, you’re better off using a revenue multiple (EV/Rev) to get a sense of valuation relative to the scale of the business.

Even for profitable companies, watch for earnings that are inflated short-term by one-time expenses, heavy SBC or accounting changes.

P/E also doesn’t reflect capital structure!

Two companies can have the exact same P/E, but one might be loaded with debt generating those earnings while the other is debt-free.

Always ask : are these earnings real, recurring and representative? And what is the strength of the balance sheet?

The Balancing Act of Putting it All Together

It’s never just one metric. It’s the interplay between all of them.

When I look at a stock, I’m running through a mental back and forth:

• What’s the TTM P/E?

• How does that compare to the forward multiple?

• What does the last two-quarter run rate tell me?

• Is the market pricing in acceleration or deceleration of growth?

• Is the PEG ratio telling me I’m paying a fair price for growth?

• Where is the P/E relative to the historical range and sector peers?

• What catalysts could drive multiple expansion?

• What headwinds could compress the multiple?

When you layer these frameworks on top of each other, a picture emerges.

Maybe the run rate P/E is compressing, the forward P/E looks reasonable, the PEG is below 1, and the stock is near the bottom of its historical valuation range while the business is inflecting with tailwinds ahead.

—> That’s what a compelling setup looks like.

This balancing act is what real fundamental analysis looks like, and once you get good at it, you’ll be able to run that checklist above in 15min, especially with the power of AI these days.

It’s not about memorizing formulas.

It’s about developing the instinct to cross-reference multiple signals and form a novel view.

Takeaway This

The P/E ratio is a starting point.

Use the TTM to orient yourself, the forward P/E to understand what the market expects, the run-rate to gauge the business right now and the PEG to contextualize the growth.

Study the historical range for expansion and contraction signals and compare multiple with relevant peers.

Layer these frameworks.

Some of these ratios won’t work for certain businesses and you’ll have to use other analyses’ in your mental toolkit to evaluate the business. Just remember to think.

That’s the toolkit. Now go use it!

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

You may have noticed that energy drink stocks are still on a tear.

• Red Bull sold a record ~14B cans in 2025 (+10.2% YoY), with global sales jumping 8.6% to about $14B+.

• Monster had a banner Q4 as well with net sales rising 17.6% to $2.1B, the first time the quarter topped $2B.

The whole category is booming, which makes the backdrop for the Celsius story much more interesting, as the 3rd biggest player now competes in a much fiercer landscape.

Subscribe now

Q4 & FY2025 Earnings

Celsius’ Q4 & FY2025 earnings were very strong:

• Q4 revenue of $721.6 M (+117% YoY)

• FY2025 sales of $2.5B (+86% YoY)

• Q4 Adj. EBITDA doubled to $134.1M (+113% YoY)

• FY2025 Adj. EBITDA was $619.6M (+142% YoY)

• Adj. EPS of $0.26 in Q4 (+86% YoY)

• Adj. EPS of $1.34 for FY2025 (+91% YoY)

In short, Celsius exceeded expectations on topline growth, EBITDA and EPS across the board, except for gross margins.

Gross margins were 47.4% in Q4 (down 2.8% YoY) due to integration costs, including Rockstar’s lower-margin mix, tariffs and one-time Pepsi distribution transition fees. However, full year gross margins finished at 50%+ (roughly flat vs. 2024).

Adj. EBITDA margin expanded massively with scale this last report — as Celsius transitions to a multi-brand business, the operating leverage is showing up in the financials.

Management also on the earnings call that as the remaining integration work completes for Alani and Rockstar, margins should rebound to low-50s% in 2026.

Celh Q4 2025 Investor Presentation Final Version

1.53MB ∙ PDF file

Download

Download

The Core Celsius Brand

I drew this up as a major risk in my last post on Celsius.

Investors have been worried about the legacy Celsius brand deteriorating with slower growth YoY.

Based on the latest earnings report, the underlying Celsius brand is still growing.

For Q4, Celsius legacy brand sales were down (8%). This may seem very weak, but U.S. tracked retail sales for the legacy brand actually increased 12.8% during Q4.

The gap between the 8% revenue decline and 12.8% retail growth was attributed to “integration-related timing dynamics” and the sequencing of orders from Pepsi…

Moving Alani Nu into the Pepsi system was a majority contributor in the temporary noise and order misalignments for the legacy brand.

The legacy brand did grow revenue 7.5% for FY2025 and maintained their 10.9% dollar share (essentially market share) of the U.S. energy drink category!

In addition to that, Circana tracker data shows ACV (All Commodity Volume) for the legacy brand has reached nearly 98.5%. Total business ACV reached 99.5%.

—> Essentially, this means that if you walk into almost any store in the U.S. that sells energy drinks, you are very likely to find a Celsius on the shelf.

That is very impressive, but does show us that revenue growth from improved distribution from the legacy brand will be much tougher, as they are already everywhere. Further growth will have to come through more shelf space with existing wholesalers, improved marketing and current customer wallet sell-through.

The Company also guided for a projected 17% increased shelf space for the legacy brand in 2026, which bolsters the foundation for the legacy brand going forward.

Legacy Celsius growth is certainly moderating from its hyper-growth days, but it’s doing so at very high volume and under complex integration conditions, so this feels reasonable and makes me comfortable with the performance.

Alani Nu Integration

This was the main highlight for the earnings report.

Alani Nu added about ~$1B of revenue for 2025, including a record ~$370M in Q4.

YoY retail sales for Alani Nu soared +101% in 2025, driven by aggressive distribution and consumer demand, as the brand has been successfully integrated into Pepsi’s system (not fully).

Alani also released new flavors during the period, which sold out in record time! This flavors include ‘Cherry Bomb’ and ‘Lime Slush’.

One of the craziest stats in the presentation was that the Alani brand surged to 94.2% ACV by the first week of Feb 2026, versus ~87% ACV when Pepsi took over in early Q4. That’s astonishingly successful integration.

—> More ACV means more shelf space and sales velocity.

In addition, Alani Nu's Q4 revenue of $370M (PF growth of 136%), run-rated through 2026 gets you ~$1.5B. Prior to the Alani Nu acquisition, Celsius’ Q1-25 LTM revenue was $1.33B at the time.

This will be a much larger business.

Once Alani becomes full integrated, I think we’ll see extended revenue growth through 2026, along with beneficial synergistic cost realization.

Management noted the Alani transition will be fully complete by end of Q1 2026, which includes supply chain, back office and commercial operations

Rockstar Integration Continues

As a reminder, Rockstar rights for the U.S. was acquired in late August 2025 for $585M. This remains the weakest link of the business.

Rockstar contributed $45M for Q4, and its underlying U.S. sales were down (~10%) in Q4 and (~11%) in FY2025.

Honestly, that decline isn’t surprising to me, as the brand was already in a slump prior to the acquisition, and is now in the process of a distributor handoff.

BUT, at CAGNY, Fieldly described a two-phase strategy for the integration here:

1. Stabilize by sharpening positioning, improving execution and focusing on the SKU set

2. Then rebuild by modernizing the brand for today's consumer

—> Management framed this rebuild of the brand as a "handful of years,“ so this transition will be slow

Based on the earnings report, Pepsi is on track to fully transition Rockstar into its system by this summer and project annual revenue post-integration is ~$250M.

We don’t have evidence of the recovery yet, but the integration is proceeding as planned and we’ll be watching to see if the retail declines slow… and eventually show signs of true reversal.

A successful turnaround would allow Celsius to optimize the portfolio, streamline sales teams and leverage Pepsi's network to drive further synergies.

CAGNY 2026 Strategic Foundation (1 week before earnings)

On February 19, leadership presented at the Consumer Analyst Group of New York (CAGNY) conference, revealing key data on portfolio scale and distribution, which actually caused an 8-10% jump in AH trading.

The most impactful data points from CAGNY were distribution-related:

• Shelf Space: Expanded by over 25%, with an additional 17% increase forecasted for the legacy Celsius brand and a 100%+ increase for Alani Nu during the upcoming spring reset (when stores refresh their products and Alani supply chain distribution becomes fully integrated), primarily in convenience stores.

• Market Reach: The fulsome portfolio now sells through more than 250,000 retail outlets at 99.5% ACV (already talked about this)

• Consumer Engagement: 52% of repeat consumers purchase products 5 or more times. This shows brand loyalty and a sticky customer base. The brand is expanding into new occasions as well, with 37% of consumers now drinking Celsius with meals and 33% in social settings.

Celsius Cagny Presentation 2026 Final

7.53MB ∙ PDF file

Download

Download

The Aggregated Portfolio

Celsius is fast becoming a multi-brand player.

As of Q4 2025 the combined Celsius/Alani/Rockstar brands command roughly 20% of the U.S. energy drink category by dollar share (+8% YoY).

This combination of brands is the #3 portfolio behind Red Bull and Monster.

Each brand occupies its own niche as well:

• Celsius hits the health/fitness consumers

• Alani Nu resonates with females/premium functional space

• Rockstar covers the traditional energy segment (maybe a bit more gamer’ish / adventure oriented)

This multi-brand approach allows the Company to go to retailer and say “We have a shelf solution for every energy drink segment.”

This is a major advantage over its competitors Monster and Red Bull.

I also think one of the biggest growth categories within the energy drink market is the zero-sugar segment, and the Celsius (legacy) brand, Alani Nu and Rockstar collectively drove ~33% of the zero-sugar energy sub-category growth in 2025. This should continue through 2026.

Wall Street’s Upgrade Cycle

Post-earnings reactions were overwhelmingly positive, led by BofA’s upgrade from Underperform to Buy, raising their price target to $65.

Broader Analyst Wave:

• B. Riley: Raised to $85

• JP Morgan: Raised to $77

• Needham: Raised to $75

• UBS: Raised to $72

• Piper Sandler: Raised to $65

• Deutsche Bank: Raised to $56

Consensus on Celsius now stands at a Strong Buy with an average price target of $64–68, implying 20–30% upside from where the stock sits today.

What’s Next?

What I’m watching for is:

• Margin stabilization: We’ll want to see gross margins tick back up now that integration and distribution shifts are nearly finished. Synergies should be realized as SG&A leverage continues.

• Legacy organic growth: I want to see continued, stable growth from the core brand, with no signs of deceleration.

• Rockstar’s integration progress: I want to see stabilized Rockstar sales and ACV lift as it comes fully onto Pepsi’s system.

• International momentum. This is a huge growth vector, with the Company having almost no exposure abroad. The international business grew ~24% this past year to ~$92.8M, but I want to see this growth accelerate. If Celsius can replicate the U.S. growth playbook abroad, that’s huge long-term upside.

At the end of my last write-up on Celsius, I mentioned:

“I am definitely formulating a plan for selling these shares within the next several months unless organic growth accelerates, integration synergies are successful, distribution is seamless (doesn’t have to be perfect), and the business continues to execute on their core offering.”

This last earnings report certainly got me more comfortable with my current ownership in the business, and I plan on holding these shares for most of 2026.

I would still say I am cautiously optimistic, but this quarter’s execution definitely puts Celsius right on the path we mapped out:

• The base Celsius brand stayed strong enough

• Alani is integrating on schedule

• The setup is laid out for Rockstar to follow

• International growth is getting stronger

In addition to this, the current forward P/E of ~28-30x (well below its 5-year median of ~77x) illustrates a reasonable valuation for a company growing revenue 85%+ with a path to mid-50s gross margins with one of the most popular energy drink brands for younger consumers.

While execution risks regarding inventory timing and stabilization remain, the risk/reward profile has shifted meaningfully in favor of shareholders heading into the remainder of 2026.

If the team stays focused and avoids any more negative surprises, the multi-brand thesis story looks alive and well.

Share

Disclaimer: The information provided in this publication is for informational and educational purposes only and does not constitute investment, financial, or other professional advice. ThePrivatePublicInvestor and its authors are not registered investment advisors or broker-dealers. All opinions expressed reflect personal views as of the date published and are subject to change without notice. While efforts are made to ensure accuracy, no guarantee of completeness or reliability is given. Past performance is not indicative of future results. The author may hold positions in securities discussed. Use of this content is at your own risk.

I transferred $300 into Coinbase and bought $300 of Bitcoin and Ethereum.

I was so excited. I thought I was going to be rich.

But no. One year later, Bitcoin was trading at ~$3,700/coin and Ethereum materially lower. I quit. I concluded that crypto was dumb and just a gambling addition.

But later in 2020 around Covid, Bitcoin exploded again. Then fell. Then exploded again in mid-2023/early-mid 2024.

I wasn’t invested in either of those 2 runs… and it made me very angry! Although, I also realized I was more of a stock investor than a crypto trader.

I started to realize Bitcoin traded in cycles. Lately, I’ve read into the supply/demand patterns and delved into the world of Stablecoins and Tokenized Deposits.

There is actually material utility within crypto and Blockchain technology as a whole.

^That is objective, not an opinion.

I want to be part of this era of crypto/blockchain adoption and believe, apart from outright owning crypto, that Coinbase will be a direct beneficiary.

Subscribe now

Executive Summary

Coinbase is no longer a crypto casino (revenue is no longer driven solely on fees from trading volume).

It hasn’t been a casino for a while now, but the market still treats it like one and that’s where the opportunity sits.

Coinbase is down over 60% from its $400+ all-time high last July. The bears have had their days, and frankly, some of their arguments are valid.

Crypto prices are soft. Q4 showed a $667 million GAAP net loss for Coinbase.

Sentiment is ugly and could get worse.

But, while everyone is staring at Bitcoin’s price chart over the past years, Coinbase has been rebuilding itself into a regulated financial infrastructure company; what the internet is calling “a tollbooth” on the rails that institutional money is increasingly forced to travel through.

Assets on the platform have grown 3x over the past three years, with Coinbase now storing over 12% of all crypto globally.

In 2025, total revenue grew 9% to $7.2B despite a soft Q4, with subscription and services hitting $2.8B, up 23% YoY and now 41% of total topline revenue.

Transaction revenue is still $4.1B and very cyclical (floats with trading volume), but the other pieces — USDC, cold storage custody, Coinbase One and derivatives are what actually drive the valuation debate now.

The question for this investment opportunity isn’t whether crypto is going to have another bad quarter. It probably will.

The question is whether Coinbase has built enough durable revenue to survive it, whether it can withstand the multitude of regulatory changes and statues going through government, and whether the long-term structural tailwinds via stablecoins, tokenized assets and regulatory clarity can be materialized and implemented successfully within the economy.

Let’s dive in.

Share

As always, I would really appreciate if you shared this post and subscribed to ThePrivatePublicInvestor, as I try my hardest each week to give you custom, in-depth analysis on stock investments, market insights and portfolio strategies.

With a subscription to the ThePrivatePublicInvestor, you will receive insight into my personal portfolio, along with each position I own and the related weighting, along with my personal watchlist, custom templates for valuation and modeling (coming soon), additional in-depth analysis on my portfolio holdings, decision-making investment guides, and the personal chat feature for the community.

As this channel grows, paid subscriptions will start to be incorporated, so subscribe early and stay subscribed to receive ‘founding member’ pricing and exclusive benefits.

With that, enjoy this piece and let me know if you have any questions!

An Updated View on How Coinbase Makes Money

Before we go further, it's worth grounding ourselves in how this business actually makes money. There are really two buckets of revenue.

In summary:

• 59% of the business is transaction based

• 41% of the business is subscription / recurring based

—> If you are already familiar with the business model as of late, feel free to skip this section

Bucket 1: Transaction Revenue ($4.06B in 2025, ~59%)

This is the legacy business, which essentially can be boiled down into fees charged when someone trades crypto on the platform. It breaks down into three pieces:

• Consumer trading ($3.32B): This is you or me buying Bitcoin on our phone. Coinbase charges a % spread on each trade, typically around 1-2% for simple trades. This is still the single largest revenue line, but it’s the most volatile. When crypto sentiment dies, this number can fall off a cliff within weeks.

• Institutional trading ($480M): Hedge funds, asset managers and other institutional corporations trading in large amounts. The fees are much lower per trade (think 0.05-0.06% of volume) because they are enormous orders.

  • This line item surged in 2025 due to the Deribit acquisition, which instantly made Coinbase the global leader in crypto options and derivatives. Deribit hit all-time high revenue in Q4.

• Other transaction revenue ($253M): These are fees from instant transfers (when you move money to your bank account quickly), Base network transaction fees and payment processing.

Note: Consumer trading margins have been trending down as competition from Robinhood, Kraken and others intensifies — this bucket will always move with crypto sentiment

Bucket 2: Subscription & Services Revenue ($2.83B in 2025, ~41%)

This is the new Coinbase — the part the market undervalues, in my opinion. These are revenue streams tied to assets sitting on the platform, not people actively trading. It breaks down into four pieces:

• Stablecoin revenue ($1.35B): Through the partnership with Circle (issuer of USDC), Coinbase earns a share of the interest dollars generated by the reserves backing USDC, which are parked in T-bills and cash equivalents (essentially a revenue-sharing agreement).

  • Think of it like this: every dollar of USDC sitting on Coinbase is effectively a deposit earning Treasury yields, and Coinbase gets a cut of that rate

• Blockchain Rewards / Staking ($677M): When you “stake” your Ethereum or Solana through Coinbase, the network pays you rewards for helping validate transactions. Coinbase takes a 25-35% commission on those rewards. (I honestly don’t know the exact nature of how this technically works in the blockchain system).

  • This revenue is denominated in crypto, so when ETH or SOL prices drop, the dollar value of rewards drops too. The nature of this actually caused a decline recently on their P&L in Q4

• Interest and Finance Income ($247M): Interest earned on customer cash balances held at banks + income from Coinbase’s institutional lending business.

• Other Subscription & Services ($555M): This comprises of Custodial fees (charged to institutions like BlackRock for storing crypto assets), Coinbase One subscription revenue (paid platform subscribers) and other platform services.

  • Coinbase holds over 80% of U.S. Bitcoin and Ethereum ETF assets in custody

“Bucket 3”: Other Revenue ($298M in 2025)

This consists of corporate interest income and other miscellaneous items. This is not a driver of the thesis but worth noting, as $11.3B in cash generates meaningful interest income on its own.

The flywheel in the shareholder letter describes the the business pretty well:

Build trust → assets concentrate on Coinbase → users and institutions adopt more products around those assets (staking, borrowing, trading, payments) → revenue grows and is reinvested back into better products and compliance

The Business Model Shift

If you asked me about Coinbase in 2021, I would’ve told you it was a leveraged bet on retail trading crypto euphoria.

• Revenue went up 80% when crypto pumped and fell off a cliff when it crashed. That was the old Coinbase!

—> The new Coinbase looks meaningfully different

In 2025, the company generated $7.18B in total revenue, with subscription revenue making up a much larger portion of sales YoY.

This subscription number is 5.5x higher than the 2021 cycle peak!

Inside that subscription bucket, the standout revenue piece is stablecoin revenue, which produced $1.35B in 2025, up 48% YoY. Stablecoin revenue alone now represents ~20% of Coinbase’s total revenue.

Note: In the last period, the ‘Other Subscription & Services’ line item and ‘Interest & Finance Income’ line items have been lumped together for reporting standards, so those 2 bar graphs may have slightly off numbers… but you get the point

The reason this breakdown matters is that it tells you what kind of company Coinbase is becoming versus what it was.

If you model forward several years and assume subscription revenue grows at 15-20% annually while transaction revenue grows steadily or stays flat, you get to a business where the majority of revenue is recurring and asset-based rather than trading-dependent.

Now that is a business that I want to own!

Small caveat: This isn’t true “SaaS” though. If ETH drops 50%, the dollar value of staking rewards, custody fees and even some USDC flows will fall too, even if nobody leaves Coinbase’ platform/ecosystem. I like to think of this as a more stable stream of revenue.

In case you missed it, here are some of the best investing insights for 2026:

• 5 Stocks Worth Your Attention in 2026

• Adobe & the AI Pivot

• Deep Dive on AMD

Stablecoin Utility is an Asset to Coinbase

If you’re going to own Coinbase, you have to understand stablecoins because this is a big part of where the thesis lives or dies in my eyes.

Read more