
AppLovin's growth runs on its models getting better, and Q2 FY2026 showed how tight that leash is: the first revenue miss in five quarters, then the next model uplift landed and Q3 reaccelerated. How much of Q3's reacceleration is the new model versus normal seasonality?
Rising spend for twelve straight quarters came from models predicting buyers better, not new customers. In Q2 FY2026 fewer usable lifts landed, revenue came in under analyst expectations, and management owned it rather than blaming demand. How many quarters in the last three years had no usable model uplift?
Q2 is the quiet stretch for shopping advertisers, yet this eighteen-month-old business set a record for spend, above its Q4 peak. Growth comes from existing customers spending more, not from new sign-ups — harder to win and slower to lose. What share of consumer ad spend comes from customers signed up a year ago?
Better models cost more to train and serve, and in Q2 FY2026 that spending rose as revenue fell short: about 70 cents of each extra revenue dollar reached profit. The cost now lands a quarter before the sales it helps create.
| Politician | Transaction | Amount | Traded | Filed |
|---|---|---|---|---|
Donald Trump Republican · Executive | Purchase | ≥$15K | Aug 28 | Oct 8 |
Donald Trump Republican · Executive | Sale | ≥$1K | Aug 24 | Oct 8 |
Donald Trump Republican · Executive | Sale | ≥$250K | Aug 21 | Oct 8 |
Donald Trump Republican · Executive | Purchase | ≥$1K | Aug 10 | Oct 8 |
Donald Trump Republican · Executive | Sale | ≥$1K | Jul 23 | Sep 22 |
Congressional trading data provided by QuiverQuant
How AppLovin turned $5.5B of revenue into $4.2B of operating income.
| Revenue | ||
|---|---|---|
| Reportable Segment | $5.5B | |
| Total revenue | $5.5B | |
| Gross profit | $4.8B | 87.9% |
| Operating income | $4.2B | 75.8% |