AppLovin Q2 revenue grows 53% to $1.92B, stock drops 21%

AppLovin's Q2 print landed as one of the sharpest reactions of this adtech earnings season. Revenue of $1.92 billion grew about 53% year over year and adjusted EBITDA hit $1.61 billion at an 83.9% margin, numbers that would flatter almost any business, but which came in fractionally below Wall Street's roughly $1.94 billion and $1.64 billion consensus expectations. The market's reply, as covered by TechCrunch and other outlets, was to send the stock down 21% to $331.57 immediately after reporting, with after-hours lows touching a 28% decline.

The reason matters more than the delta. CEO Adam Foroughi told investors that "our pace of meaningful model improvement was lighter than normal during the quarter, and the next step up in model performance landed just after quarter end." That is a striking admission from a company whose entire premium multiple is priced on model-driven adtech performance. Q3 guidance of about $2.07 billion at the midpoint against a $2.08 billion estimate is strong on any absolute basis, but it is not the flawless cadence a $141 billion market cap had been paying for.

Not every read was negative. Foroughi flagged that e-commerce customers achieved record spending, up 28% year-over-year, and CFO Matt Stumpf disclosed in commentary reported by Stocktwits that "the SEC has recently advised us that it concluded its inquiry with no recommended action." The probe had examined data-collection practices and alleged unauthorized "fingerprinting" techniques, so lifting that cloud is worth real money to a stock already down 38% year to date.

The honest caveat is that the reporting does not tell you what the post-quarter model step-up actually does in production, whether the e-commerce vertical scales fast enough to offset any lull in gaming ads, or whether the SEC's no-action closure ends related private exposure or only the regulator's piece. Those blanks are where the next quarter gets decided.

The setup for buyers on the dip is not unreasonable if you take management at their word: the AI improvement they wanted arrived just after the quarter closed, the SEC overhang has lifted, and e-commerce is compounding. The setup for skeptics is equally clear: the market had been paying for perfect model iteration, and one bumpy quarter proves that assumption is now in doubt.