Philip A. Fisher | February 22, 2026 Period: Q4 FY25 (quarter ended December 31, 2025)
This is my first formal analysis of AppLovin. My prior was informed by the company's public record: a mobile advertising network that, through its AXON 2.0 machine learning system, achieved something genuinely unusual in the technology business — a step-change improvement in product efficacy that translated almost immediately into financial results of extraordinary quality. The reservation I carried in was the Muddy Waters short report (late 2024) and the structural question of whether a company that owns both the advertising platform and inventory within that platform has a fundamental conflict of interest. I wanted management to address that directly.
I will begin here, as I always do. The financial statements tell us what happened; scuttlebutt tells us why and whether it will continue.
Customer sentiment — mobile gaming: The picture from the gaming advertiser community is unambiguously positive. Mobile gaming studios — from the largest (Zynga, Scopely) down to hypercasual developers — report that AXON 2.0 outperformed legacy systems on install efficiency by substantial margins, with some reporting 30-50% improvements in cost-per-qualified-user. The product has a "set it and forget it" quality for smaller studios that lack sophisticated user acquisition teams — which is precisely the stickiest kind of product advantage. The negative: opacity. Advertisers cannot fully audit the targeting logic. In my experience, this is actually a scuttlebutt positive when accompanied by results — advertisers do not complain about black boxes that work.
Customer sentiment — e-commerce: Here the picture is cloudier, and appropriately so. AppLovin only announced its e-commerce push in late 2024. Early beta feedback was cautiously positive on reach and CPM pricing. But the fundamental challenge is clear from any honest assessment of the competitive dynamics: e-commerce advertisers have Meta, Google, TikTok, and Amazon as alternatives. Gaming advertisers, post-Apple's App Tracking Transparency changes that disadvantaged Meta's gaming inventory, had far fewer credible alternatives. AppLovin's data moat — 1.4 billion daily active devices in gaming apps — is rich for behavioral signals but largely unproven for purchase-intent signals. The 57% go-live rate management cited on the call and the 30-day LTV breakeven metric are encouraging, but they are early evidence from a self-selected beta cohort, not a broadly representative sample.
Employees: AppLovin's Glassdoor ratings have been consistently above average for the adtech sector. The common themes — competitive compensation, engineering-driven culture, fast decision-making from leadership — describe a company that attracts and retains capable builders. The high-performance culling culture and rapid strategic pivots generate negative reviews but are consistent with what I would expect of a management team focused on results over comfort. I found no evidence of meaningful talent flight.
Competition: The most credible competitive threat in the record is Moloco — a private, machine-learning-native demand-side platform growing among mid-tier gaming companies. Unity's competing ad network has been substantially weakened by operational turbulence since 2023, likely a net beneficiary for AppLovin through share shift. "CloudX" — the name cited as a catalyst for the stock's 41% YTD decline — does not appear in established competitive intelligence as a product of record. Foroughi addressed this directly on the Q4 call, describing what I would characterize as a competitive dismissal grounded in specific product differentiation claims rather than vague corporate posturing.
Management reputation: Adam Foroughi has become notably more public through 2025. His framing — AppLovin as an AI company first, advertising infrastructure second — is disciplined and consistent. His direct rebuttal of competitive concerns on the Q4 call, and his comparison of e-commerce to gaming as a sequential market expansion, reflects the kind of long-range thinking I look for in Point 4. He is dismissive of Unity, silent on Moloco (which is appropriate competitive hygiene), and consistently frames the TAM expansion from gaming (50B)toallmobileadvertising(400B+) as the governing narrative.
Short sellers: The Muddy Waters report (late 2024) raised two concerns: first, that AXON's performance metrics were partially inflated through incentivized installs; second, a structural conflict of interest — AppLovin owns apps that run AppLovin's own ads. Management denied the allegations and pointed to continued advertiser retention. By February 2026, the short thesis has not been vindicated by financial results. Revenue and margins continued to expand. The structural conflict concern is real and worth monitoring, but it is not a smoking gun. I have seen many businesses with vertical integration that was initially characterized as a conflict — and later recognized as an integrated value chain. The test is whether the advertising results hold for independent advertisers. They appear to.
OpenAI speculation: Unconfirmed. Not actionable. I do not analyze rumors.
| Metric | Q4 FY25 | Q3 FY25 | Q4 FY24 | YoY |
|---|---|---|---|---|
| Revenue | $1,658M | $1,198M (adj.) | $999M | +66% |
| Software Platform Rev | $1,476M | — | $835M | +77% |
| EBITDA [Non-GAAP] | $1,399M | — | — | — |
| EBITDA Margin [Non-GAAP] | 84.4% | — | — | — |
| GAAP Op Margin | 76.9% | — | — | — |
| FCF | $1,309M | — | — | — |
| FCF Margin | 78.9% | — | — | — |
FY25 Full Year:
Q1 FY26 Guidance:
The absolute numbers are exceptional by any standard of the technology industry. An 84% EBITDA margin on a $1.7B quarterly revenue run rate is not a product of financial engineering — it is the signature of a software business with minimal marginal cost of delivery and a pricing position that reflects genuine advertiser dependence on the product. The 95% incremental EBITDA flow-through Atlas documented means that each additional revenue dollar drops almost entirely to cash. This is the economics of a business "fortunate because it is able."
The deceleration from 66% YoY growth in Q4 to implied ~52% in Q1 FY26 merits attention. In isolation, deceleration from extraordinary growth is not alarming — it is arithmetic. The governing question is whether the deceleration reflects market saturation or addressable market expansion into e-commerce. I believe it is the latter, but it is too early to confirm.
Point 1 — Does the company have products or services with sufficient market potential to make possible a sizable increase in sales for at least several years? Yes, with high conviction. The gaming advertising market (50BTAM)iswell − penetrated; thee − commerceadvertisingmarket(400B+ TAM) is nascent for AppLovin. The company has demonstrated the playbook works in gaming; the question is whether AXON's model generalizes. Management's 30-day LTV breakeven evidence in e-commerce beta is promising. Pass.
Point 2 — Does the management have a determination to continue to develop products or processes that will still further increase total sales potentials when the growth potentials of currently attractive product lines have largely been exploited? The e-commerce pivot is exactly this — a planned transition to the next growth leg before gaming saturates. The gen AI creative tools pilot (100+ customers) and video model launch are further evidence of deliberate expansion into adjacent capability. Pass.
Point 3 — How effective are the company's research and development efforts in relation to its size? AppLovin's "R&D" is its machine learning infrastructure — AXON 2.0 and its successors. The 66% YoY revenue growth achieved without a proportionate increase in headcount or cost base suggests the model is compounding its own efficacy. EBITDA/employee ratio of approximately $5.6M is evidence of this leverage. The gen AI creative pilot represents the next product layer above the bidding engine. Pass.
Point 4 — Does the company have an above-average sales organization? This is the most nuanced point. AppLovin has historically relied on product performance rather than a traditional sales force — the AXON model sells itself when it works. The 57% go-live rate in e-commerce beta suggests that adoption friction exists; not every prospective advertiser converts. The "fantastic" prospecting product Foroughi referenced on the call, and the planned GA for H1 2026, will be the test of whether AppLovin can acquire e-commerce advertisers at scale with a lighter sales motion. Somewhat uncertain but directionally favorable. Pass with watch.
Point 5 — Does the company have a worthwhile profit margin? 84% EBITDA margin. The question answers itself. Pass emphatically.
Point 6 — What is the company doing to maintain or improve profit margins? The margin is maintained by operational leverage — revenue scaling faster than cost, and the AI model improving without proportionate headcount growth. The Apps divestiture in Q1 FY25 eliminated lower-margin consumer business. The incremental margin on new e-commerce revenue will be the test going forward; no reason yet to expect deterioration given the same software delivery infrastructure services the new vertical. Pass.
Point 7 — Does the company have outstanding labor relations? Employee reviews are above average. No material talent flight documented. The engineering-driven, high-performance culture attracts builders. Pass.
Point 8 — Does the company have outstanding executive relations? The company runs lean. Foroughi's direct rebuttal style on the Q4 call — "real disconnect between market sentiment and reality" — reflects a management team that communicates with confidence rather than hedging. No documented executive departures of concern. Pass.
Point 9 — Does the company have depth to its management? This is an area of genuine uncertainty. AppLovin is Foroughi's company in a way that creates key-man concentration. The public record does not reveal a deep executive bench. This is a concern worth monitoring. Partial pass — watch.
Point 10 — How good are the company's cost analysis and accounting controls? FCF of 78.9% of revenue, virtually no divergence between reported EBITDA and cash generation. SBC at 4.9% of revenue is responsible for an AI company. These are signals of disciplined financial management. Pass.
Point 11 — Are there other aspects of the business, somewhat peculiar to the industry involved, that will give the investor important clues as to how outstanding the company may be in relation to its competition? Yes: the data flywheel. AppLovin's 1.4B daily active device signals feed AXON's model, which improves targeting, which improves results, which attracts more advertisers, which funds more app publisher relationships, which produces more signal. This is the structural moat — not a brand or a patent, but a self-reinforcing data advantage that compounds over time. Competitors cannot buy this moat; they must build it, which takes years. Pass.
Point 12 — Does the company have a short-range or long-range outlook in regard to profits? The Apps divestiture — sacrificing near-term revenue to simplify the model — is a long-range decision. The e-commerce expansion, with its investment phase before full monetization, is a long-range decision. The buyback at current prices reflects confidence in multi-year value creation. Pass.
Point 13 — In the foreseeable future will the growth of the company require sufficient equity financing so that the large number of shares then outstanding will largely cancel the existing stockholders' benefits from this anticipated growth? Net debt of $1,026M against $4.5B annual EBITDA is immaterial. The company is generating $4B in annual free cash flow and returning $2.5B+ to shareholders via buybacks. No dilutive equity issuance needed. Pass.
Point 14 — Does management talk freely to investors about its affairs when things are going well but "clam up" when troubles and disappointments occur? The Q4 call is instructive. Foroughi addressed the CloudX competition narrative directly rather than avoiding it. He was specific about e-commerce progress metrics (30-day LTV breakeven, 57% go-live) rather than vague. He characterized market sentiment as disconnected from reality — a confident, frank assessment. Management has not "clammed up" in the face of short-seller reports or stock price declines. Pass.
Point 15 — Does the company have a management of unquestionable integrity? The Muddy Waters conflict-of-interest allegation — AppLovin owns apps that run AppLovin's own ads — is the only integrity question in the record. The allegation has not been confirmed or elaborated into demonstrated harm to independent advertisers. The Apps business has been divested, which removes the most direct form of the conflict. The continuing ownership of Max (the mediation layer) and the advertising platform creates ongoing vertical integration that warrants monitoring. I do not conclude integrity failure, but I note the question has not been definitively resolved. Pass with watch — this is the single point I hold open.
Summary: 13 clear passes, 2 conditional passes (management depth, Point 15 integrity overhang). No failures.
AppLovin's growth arc is the story of a well-constructed machine learning system applied sequentially to expanding markets.
Phase 1 (complete): Gaming UA. AXON 2.0 achieved product-market fit, drove 66-70% YoY revenue growth, and built the data flywheel. This phase is largely exploited — gaming remains a large market but AppLovin's penetration is deep enough that incremental share gains have diminishing returns.
Phase 2 (in progress): E-commerce advertising. The playbook is identical: apply AXON's bidding and targeting intelligence to a new advertiser vertical. The market is 8x larger than gaming. Early signals are positive. GA is H1 2026 — the next two to three quarters will be definitive.
Phase 3 (early): Gen AI creative tools. Foroughi's "fantastic" creative product pilot with 100+ customers represents a move up the value chain — from distributing ads to generating them. If successful, this both raises switching costs and creates an additional revenue layer on top of the media buying business.
This is a company "fortunate because it is able" in Fisher's full sense. The management team has deliberately engineered each successive expansion rather than simply riding the gaming industry's growth. The data flywheel, once established, becomes the platform for each phase.
Adam Foroughi earns high marks on the Fisher dimensions that matter most: long-range thinking, candor under pressure, product-first orientation, and demonstrated willingness to make structural sacrifices (Apps divestiture) for long-term clarity. The Q4 call performance — direct rebuttals, specific metrics, confident framing — is the behavior of a manager who believes in his business and is willing to say so plainly.
The concern I hold is management depth. Fisher's Point 9 is about what happens to a business when the founding CEO is no longer present. AppLovin has not demonstrated a deep executive layer that could sustain the culture and strategy independent of Foroughi. This is worth monitoring but not a disqualifying condition at this stage.
Atlas scored this at Conviction 4/5 with "compelling setup" framing. I agree with the broad assessment. Where I add nuance:
Agreement: The financial quality is exceptional. The 95% incremental margin flow-through, the e-commerce conversion headroom (1%→5% of visits), the data flywheel moat — Atlas correctly identified these as the central bullish arguments.
Divergence — emphasis: Atlas framed this as a valuation opportunity (41% selloff = buying opportunity). I am more cautious about valuation-driven framing. My framework does not set price targets. What I care about is whether the business fundamentals remain intact or are improving. They are. The selloff, in my judgment, reflects a market extrapolating a competitive threat (CloudX/AI competition) that has not yet appeared in any financial result. That is a temporary mispricing if the business fundamentals hold.
Divergence — Point 15: Atlas did not emphasize the Muddy Waters conflict-of-interest concern. I hold it as an open question. The Apps divestiture reduces but does not eliminate the structural conflict.
Divergence — e-commerce: Atlas was bullish on the expansion. I am bullish but more conditional — I want to see Q2 FY26 data showing e-commerce advertiser retention and ROAS outcomes before elevating my conviction.
AppLovin passes the Fifteen Points on thirteen clear criteria with two conditional watches. The financial performance in Q4 FY25 was the best in the company's history: $1,658M revenue at 84% EBITDA margins, generating $1.3B in free cash in a single quarter. The management team has earned trust through consistent execution, transparent communication, and structurally sound capital allocation.
The 41% YTD stock decline preceding this earnings report reflected competitive fear, not fundamental deterioration. Foroughi's direct rebuttal of those fears, combined with Q1 FY26 guidance of $1.76B at maintained margins, suggests the fears were premature.
The e-commerce expansion is the governing question for the next twelve to eighteen months. If AXON generalizes to the e-commerce vertical with even a fraction of the efficacy it demonstrated in gaming, the financial results will be substantially larger than present estimates. If it does not, growth will decelerate toward the mid-20s as the gaming market matures.
I hold this as a strong buy on the fundamental merits. The business qualifies on the Fifteen Points. Management has demonstrated long-range thinking and candor. The data flywheel moat is real and compounding. The price decline was driven by narrative, not numbers.
The time to sell is, as I have always said, almost never — when a business of this quality is executing and expanding its market, the investor who sells on competitive fear alone will almost certainly regret it.
Thesis status: Initiating — Strong conviction. Key monitor: E-commerce ROAS outcomes (Q2 FY26 report). Management depth. Point 15 integrity overhang.