BLANC WM
Equity Research
Investment Memo — AppLovin (APP)
21 Sep 2026 · Confidential
RU
APP · BUY Initiating Coverage House Conviction: High

AppLovin — Punished Compounder, Largest Overhang Retired

Down ~57% from Dec-2025 highs on a stack of overhangs the house view considers largely resolved or overstated, against fundamentals that remain exceptional

APP fell from an ATH of $733.60 (Dec 2025) to ~$310 (Sep 2026) on a stacked sequence of short-seller campaigns, an SEC data-practices probe, a first-ever guidance miss in Q2, and a sector-wide AI-disruption multiple reset. The SEC closed its investigation in August with no enforcement action. Fundamentals remain exceptional — 53%+ revenue growth, 84–85% EBITDA margins, ~$863M quarterly FCF. House view: the market is still pricing legal/regulatory tail risk that has substantially cleared. Consensus targets of $500–560 imply 60–75% upside from current levels.

What happened. Four overlapping pressure points drove the drawdown: (1) short-seller reports (Fuzzy Panda, Culper, Muddy Waters) alleging AXON violated partner platform terms of service via "identifier bridging"; (2) an SEC Cyber and Emerging Technologies Unit probe into the same allegations, opened Oct 2025; (3) a Q2 2026 print that missed revenue by $16.3M and guided Q3 below consensus — AppLovin's first-ever guidance miss since IPO, which triggered a 28.7% after-hours drawdown and $40B of market-cap erasure on what was, in isolation, a marginal miss; (4) a January 2026 sector-wide de-rating of high-multiple software/AI names tied to Google's Project Genie launch and broader AI-disruption repricing.

What has since changed. The SEC closed its investigation in August 2026 with no enforcement action — the single largest legal tail risk in the bear case is now off the table. AppLovin's forward P/E has compressed from a 2025 peak near 47–88x to ~26x, closer to normalized adtech multiples, while growth (53%+ YoY) remains 4–5x that of comparable adtech names. The Q2 miss was a ~1% revenue shortfall against guidance — immaterial in magnitude, but symbolically significant as a first break in an otherwise clean beat-and-raise record.

House view on the five commonly-cited threats: the house position is that all five are low-probability or already substantially addressed by the SEC's closure, with the qualification (flagged below) that platform/ToS dependency is a live commercial risk independent of the legal question and is not reducible to zero by a regulatory clearance.

Price (19 Sep)
~$310
52w range $297.50–$745.61
Drawdown from ATH
−57%
ATH $733.60, 22 Dec 2025
Fwd P/E
~26x
vs. 47–88x in 2025
Q2 Rev Growth
+53%
YoY, EBITDA margin ~84–85%
Quarterly FCF
~$863M
Q2 2026
TipRanks Avg PT
$511
+71% implied · Strong Buy (17B/5H/0S)
I. Catalysts For Re-Rate
SEC Investigation Closed
No Action
Aug 2026: SEC's Cyber and Emerging Technologies Unit closed its year-long AXON data-practices probe with no enforcement action. This was the most credible existential threat in the bear case — four separate short-seller reports built a thesis regulators examined and declined to act on.
📊
Multiple De-Rated Faster Than Fundamentals
88x → 26x
Forward P/E has compressed to levels close to mature-adtech comps, while growth (53%+) remains far above peers. Trade Desk, by contrast, has genuinely slowing growth (~12%) — APP's Q2 wobble was a ~1% miss, not a trend break.
II. Five Threats — House View vs. Street Concern
Risk Scorecard
House position: all five overstated or resolved; one retains an open analyst flag
ThreatStreet ConcernHouse View
1. Growth Deceleration Q2 guide-miss signals a structural slowdown from 50%+ toward a mature-adtech growth rate. Groundless Q3 guide reaffirmed $2.07–2.08B at 83% EBITDA margin; Q2 miss was ~1% of guidance, management attributed it to model-timing, not demand.
2. AI/Google Project Genie Disruption AI-native game creation threatens the mobile-gaming ad ecosystem AppLovin monetizes. Groundless Thesis-level, multi-year risk with no observed impact on AppLovin's Q1–Q2 2026 metrics; a broader software-sector narrative, not AppLovin-specific.
3. Litigation Overhang Bernstein Liebhard securities class action (filed 17 Sep 2026) keeps "were the numbers real" question alive. Low Probability Filed post-SEC-clearance; house view is this is opportunistic plaintiffs'-bar activity following a stock decline, not evidence of new wrongdoing. Historically APP short-thesis allegations have not stuck.
4. Platform / ToS Data Dependency AXON's edge depends on data access from Meta, Google, Snap, TikTok, Reddit that these platforms could tighten unilaterally. Groundless (house) SEC reviewed the identifier-bridging allegations directly and took no action; company denies violation.
5. Insider Selling 194 recent insider transactions net-selling signals management doesn't see value at current levels. Low Probability Consistent with routine 10b5-1 diversification at a founder-heavy cap table post a 400%+ multi-year run; not a fresh signal distinct from ordinary-course selling.
Analyst flag on Threat 4: an SEC enforcement decline is a legal outcome, not a commercial guarantee. AppLovin's targeting edge still depends on continued data access from partner platforms that can tighten terms unilaterally at any time, independent of any regulatory finding. This is treated here as a permanent, low-but-nonzero structural risk factor rather than one retired by the SEC's closure — the house "groundless" call above reflects conviction on the legal/regulatory question specifically, not a claim that the underlying commercial dependency has disappeared.
III. Sell-Side Support
Corroborating Views
Independent sell-side and short-term-catalyst desks reach the same conclusion on the Q2 wobble

Citi (Buy, PT $600, 11 Sep 2026): "We rate AppLovin a Buy. We like the shares of APP given the firm's strong growth prospects and strong cash flow generation... In addition, we see incremental growth potential if the company is able to successfully expand beyond mobile gaming by servicing e-commerce companies." Target of $600 implies ~29x 2027E price-to-FCF — a premium to the S&P 500 NTM FCF multiple that Citi views as justified given growth and FCF generation. Citi's target sits above the $500–560 consensus band cited in the Executive Summary — i.e. Citi is above-consensus bullish.

Short-Term Catalyst Desk (Upside, added 11 Sep 2026, expires 10 Dec 2026): Directly rebuts the Q2-deceleration read underlying Threat 1 in the scorecard above — attributes the 2Q26 sequential slowdown (4% QoQ) to normal seasonality (2Q is typically the slowest quarter) and to the fact 2Q26 predated the July 2026 AXON model improvement, and argues the market is wrongly discounting the 3Q26 guide of 7–8% sequential growth as too aggressive. Longer-term view aligns with the house e-commerce-expansion thesis: e-commerce TAM exceeds gaming, existing e-commerce client ROAS is attractive, and client onboarding (organic + tuck-in M&A) could accelerate — the same multiple-expansion mechanism flagged in Section I.2 above.

TipRanks Consensus (21 Sep 2026): 22 analysts with price targets issued in the trailing 3 months — 17 Buy, 5 Hold, 0 Sell, rated Strong Buy. Average price target $511.24, range $325–$790, implying ~71% upside from the $298.59 reference price at time of pull. This independently corroborates both the $500–560 consensus band cited in the Executive Summary and sits below Citi's $600 PT — i.e. Citi is above the TipRanks average, not an outlier bear or bull relative to the broader sell-side.

William Blair (Ralph Schackart, Buy reiterated, 1 Sep 2026): "The MAX platform [is] still growing at a healthy double-digit pace despite quarterly shortfalls," with e-commerce advertising "scaling from a smaller base with strong percentage growth." Schackart frames the post-earnings pullback as long-only investors mispricing a temporary internal-execution and seasonal miss rather than competitive share loss — consistent with the house Threat-1 rebuttal. One caveat worth carrying forward: Schackart flags take-rate compression as a genuine near-term monetization headwind layered on top of the seasonal effect, distinct from the demand-growth story — this is a nuance the house scorecard's "Groundless" call on Threat 1 should hold alongside, not ignore.

FirmAnalystPrice TargetRatingUpsideDate
CitiJason Bazinet$600BUY94.77%11 Sep 26
JefferiesJames Heaney$700→$550BUY78.54%18 Aug 26
Evercore ISIRobert Cooper$630→$510BUY65.55%1 Sep 26
NeedhamBernie McTernan$500→$475BUY54.19%26 Aug 26
Morgan StanleyMatthew Cost$450BUY46.08%20 Sep 26
BenchmarkMike Hickey$500→$440BUY42.83%17 Aug 26
BTIGClark Lampen$408→$396BUY28.55%8 Sep 26
William BlairRalph SchackartBUY3 Sep 26
Piper SandlerJames Callahan$325HOLD5.50%31 Aug 26
Wells FargoAlec Brondolo$357→$325HOLD5.50%19 Aug 26
Read-through: three independent reads — Citi, the short-term catalyst desk, and the 22-analyst TipRanks panel — converge on Buy/Strong Buy with no Sell ratings represented in the TipRanks panel. The zero-Sell-rating detail is notable given the past year's short-seller campaigns and litigation overhang covered in Section II: the sell-side has had ample opportunity to turn bearish on those same facts and has not. Note the spread even among Buy-rated firms is wide ($396 BTIG to $600 Citi) and several firms (Evercore, Jefferies, Needham, Wells Fargo, Benchmark, BTIG) have cut targets in Aug–Sep 2026 while maintaining their rating — consistent with the house's own view that the Q2 miss warranted a numbers reset, not a thesis reset.
Growth-Adjusted Valuation vs. SaaS/Software Peer Set
NTM revenue growth vs. forward EV/Sales, broad software comp set

Plotted against a broad software/SaaS peer set on NTM revenue growth (x-axis) vs. forward EV/Sales (y-axis), APP sits furthest right on growth (highest in the set, ~35% NTM) while trading at ~11x forward EV/Sales — below the peer-set regression line, and a fraction of the multiple names growing a third as fast command (CRWD and NET both trade 35x+ EV/Sales at roughly two-thirds AppLovin's growth rate). APP is also flagged with a positive operating margin (>5%) in this set, unlike several higher-multiple peers with negative or thin margins. On a growth-adjusted basis this is one of the cheaper names in software, not an expensive one — consistent with the reverse-DCF finding above that the market is pricing a deceleration well below both trend and consensus.

Unit Economics — Income Statement Waterfall
Revenue-to-net-income conversion, quarterly detail
Line ItemValueMargin / Note
Revenue$1,405M+68% YoY (US $688M +62%, RoW $717M +75%)
Gross Profit$1,230M88% margin, +2pp YoY
Operating Profit$1,079M77% margin, +13pp YoY
Net Profit$836M59% margin, +7pp YoY

The 59% net margin reflects a revenue-to-cash conversion profile with very little friction between the top and bottom line: 88% gross margin, opex (S&M $49M + R&D $44M + G&A $59M) totaling only ~11% of revenue, and 13pp of operating-margin expansion YoY on top of already-high 2025 levels. This is the underlying engine behind both the revenue-per-employee comparison above and the FCF-conversion assumption in the reverse DCF — a business this efficient converting a large share of incremental revenue straight to profit is exactly the profile that makes a 15%-for-a-decade priced-in bar look conservative.

Data vintage flag: this income-statement detail is sourced to a Q3 FY25 breakdown, not the most recently reported quarter (Q2 FY26, referenced elsewhere in this memo). It is included here as a structural illustration of AppLovin's cost architecture and margin conversion mechanics, not as the latest quarterly print — Q2 FY26 actuals (53% revenue growth, 84–85% EBITDA margin per Section Exec Summary) should be used for any current-quarter modeling.
IV. Reverse DCF — What The Market Is Pricing
Implied Growth At Current Price
10-yr reverse DCF, SBC-adjusted TTM FCF $4,245M, 12% discount rate, 3% terminal growth
InputValue
Share Price (reference)$334.86
Market Cap~$112.0B
Implied FCF Growth (Yrs 1–10)15.0%
Estimated Intrinsic Value at 15% Growth$331.90 (~fair value at current price)

At the current price, the market is pricing roughly 15% annualized FCF growth for the next ten years — not 50%+, not even the 29% CAGR through 2028 that sell-side consensus currently models, and a fraction of the ~45.6% FCF CAGR AppLovin has actually delivered since 2018. The gap between the priced-in bar (15%) and both the historical run-rate (45.6%) and the forward consensus (29%) is the core of the house re-rate thesis in numeric form: this is not a story where fundamentals need to accelerate from here to justify upside — a straightforward deceleration to roughly a third of trend growth, sustained for a decade, is already more than priced.

Read against Section I. A business converting revenue to free cash flow at AppLovin's rate (84–85% EBITDA margin, ~$4.2B TTM FCF against a ~$112B cap) clearing a 15%-for-10-years bar is a low bar in absolute terms for a name with this unit economics profile. The reverse DCF is why the house treats Threats 1–2 (growth deceleration, AI disruption) as needing to be not just plausible but severe and sustained to justify current levels — a moderate, temporary slowdown (which is what Q2 was) doesn't touch the embedded assumption.

Caveat: reverse DCF states what the market is pricing, not what will happen; it is a sentiment/valuation cross-check, not an independent growth forecast. Discount rate (12%) and terminal growth (3%) assumptions materially move the implied growth figure — this is presented as a single reference case, not a range.
V. Position

Rating: BUY. Entry near current levels (~$310) against consensus targets of $500–560 implies 60–75% upside on a re-rate back toward historical adtech-growth multiples, contingent on two to three consecutive clean quarters reestablishing the beat-and-raise pattern broken in Q2. Citi's $600 PT (~29x 2027E P/FCF) sits above the consensus band. The reverse DCF shows the market pricing only 15% FCF growth for a decade against a 45.6% historical run-rate and 29% forward consensus — a wide margin of safety if growth merely normalizes rather than accelerates. Operating leverage supports the FCF-conversion assumption behind that math: AppLovin's TTM revenue per employee (~$7.7M) exceeds NVIDIA's (~$7.2M) and is well above Meta (~$3.0M), Google (~$2.3M), and Amazon (~$492K) — a lean, software-margin cost structure, not a headcount-driven one, which is what makes the 84–85% EBITDA margin durable rather than a temporary peak. Watch items: Q3 print vs. $2.07–2.08B guide and 83% EBITDA margin; net new AXON e-commerce advertiser adds; resolution timeline on the Bernstein Liebhard suit; and short interest / days-to-cover trend as a capitulation signal.

This memo reflects the house view of Blanc WM as of 21 Sep 2026 and is prepared for internal use only. It is not investment advice and does not constitute a recommendation for any specific client account. Figures are drawn from public disclosures and third-party data aggregators (company filings, sell-side estimates, Fintel, GuruFocus, MacroTrends) as of the dates cited and may be revised. Past performance is not indicative of future results.