AppLovin Corporation (NASDAQ:APP) reported second-quarter 2026 revenue growth of 53%, yet its stock fell sharply after missing revenue expectations and management’s own execution standard. The tension is not whether the advertising platform is growing; it is whether machine-learning improvements can sustain the exceptional margins and earnings trajectory implied by its valuation. In early October, analysts began questioning competing ad platforms’ progress and the reliability of new e-commerce adoption signals. This makes AppLovin’s stock less a straightforward growth story than a test of how durable its targeting advantage really is.
A Small Miss Has a Large Meaning
Second-quarter revenue reached $1.92 billion against a $1.94 billion consensus estimate, while EPS of $3.76 matched expectations. Adjusted EBITDA increased 58% to $1.61 billion, implying an adjusted EBITDA margin above 80%. Management said its important advertising-model upgrade arrived just after the quarter closed. The problem was timing rather than a reported collapse in customer demand, according to management, but the distinction matters only if later results confirm it.
AppLovin Corporation ranks fourth among the 10 best undervalued stocks to buy now according to Reddit, behind three companies with very different economics. The more interesting comparison is whether AppLovin’s extraordinary operating profitability can survive a market in which rivals increasingly adopt similar AI tools.
Incremental Compute Spending Could Help or Hurt
The company’s advertising auctions improve as its models become more effective at predicting which ads produce sales. When advertisers earn better returns, they may expand budgets, giving AppLovin Corporation more data and incremental revenue. The company guided Q3 revenue to $2.055 billion to $2.085 billion, up 46% to 48% year-over-year, alongside $1.71 billion to $1.74 billion of adjusted EBITDA and an approximately 83% margin. That guidance assumes model improvements already deployed, not hypothetical later releases.
Yet compute is not free. Management explained that higher training costs are being undertaken where incremental returns justify them. It said roughly ten cents of each incremental revenue dollar could be spent on compute. An 80%-plus adjusted EBITDA margin gives the company considerable room for investment, but also means small efficiency setbacks can have an outsized impact on investor assumptions about its operating leverage.
The expansion beyond games has its own measurement problem; how AppLovin is extending its AI platform into e-commerce advertising analyzes why a larger addressable market does not immediately translate into verified revenue. Competition is another pressure point. In October, a court declined AppLovin’s request for interim relief concerning Unity’s Ad Quality SDK, without ruling on the underlying allegations. Jefferies’ survey of mobile-game ad buyers led it to raise estimates for Unity while lowering its 2027 AppLovin revenue forecast by 5% and EBITDA estimate by 7%.
The Discount Is Real, but So Is the Premium
The Seeking Alpha data places AppLovin Corporation at 20.16 times trailing GAAP earnings and 16.76 times forward GAAP earnings, with forward non-GAAP P/E of 15.81. That last measure sits roughly 46% below the company’s five-year average of 29.17, but about 26% above the sector median of 12.52. Forward enterprise value to sales is 10.91, nearly six times the sector median of 1.88. Accordingly, a much lower earnings multiple relative to AppLovin’s history does not mean the business is inexpensive on every basis.
The consensus expects EPS growth of 60.49% in 2026, 27.30% in 2027, and 24.88% in 2028. The estimated P/E correspondingly falls from 16.76 on 2026 earnings to 13.16 on 2027 earnings and 10.54 on 2028 earnings, assuming a constant stock price and forecast delivery. The steepest growth is front-loaded, so valuation depends heavily on the bridge from rapid model-led gains to a more sustainable rate.
Free cash flow reached $863 million in Q2. Management expects full-year free cash flow conversion to normalize near 75% of adjusted EBITDA after tax and interest timing effects. The company ended the period with $3.05 billion in cash and around $3.7 billion in debt and repurchased or withheld about $551 million of stock. These are meaningful financial supports, but do not settle the question of how much customer acquisition efficiency can improve from today’s elevated level.
Institutional Conviction Faces a Near-Term Test
Q2 2026 hedge fund ownership declined to 88 from 91 funds, while short positions increased to 11.61 million shares on September 30 from 10.56 million a month earlier, equaling 3.99% of float. The positioning is not exceptionally bearish by itself, but the direction contrasts with continued strong revenue growth.
Price-target cuts further illustrate the gap between current profitability and confidence in its future trajectory. On October 12, Oppenheimer reduced its target to $490 from $560 while lowering its FY2027 revenue and adjusted EBITDA forecasts; BofA maintained Neutral but reduced its target to $325 from $400 on October 9, arguing that a durable reacceleration would be needed to justify a premium to other advertising giants. Wedbush remained Outperform at $560 from $610 on October 14 but reset estimates to a more achievable baseline.
The immediate test is therefore not simply another 50% revenue-growth quarter. It is whether the Q3 model upgrade drives measurable ad spending, consumer-advertising diversification, and cash conversion without permanently higher compute intensity. A different Reddit screen asks which stocks investors expect to rise dramatically despite substantial uncertainty, underscoring the distinction between an exciting growth narrative and the earnings needed to support its price.
READ NEXT: AppLovin (APP) Grew 53% and Got Cheaper. Is the Market Missing Something? and Jim Cramer Remains Wary of AppLovin (APP) Despite Its 44% Decline
