AppLovin’s third-quarter revenue rose 55% to $2.18 billion, and about 81 cents of each added revenue dollar turned into operating income, by our arithmetic on its results release filed with the SEC on 4 November: revenue rose about $780 million from a year earlier, and income from operations rose about $630 million.
Net income rose 72% to $1.44 billion and diluted earnings per share were $4.30. Adjusted EBITDA was $1.84 billion, an 84% margin. For the fourth quarter the company guided revenue of $2,365 million to $2,395 million and adjusted EBITDA of $1,995 million to $2,025 million, a margin of 84% to 85%.
Business Pill · Operating Leverage
An evergreen Business Engineer explainer on operating leverage. It teaches the general idea only and does not discuss AppLovin.
The key insight: AppLovin added about $780 million of third-quarter revenue from a year earlier and about $630 million of income from operations, so about 81 cents of each added revenue dollar became operating income, by our arithmetic; cost of revenue, which its 10-Q says consists primarily of datacenter costs, took about 8 cents of it.
Total costs and expenses rose $149.8 million, about 19% of the added revenue, by our arithmetic. Cost of revenue, which AppLovin’s latest 10-Q says “consists primarily of datacenter costs related mainly to third-party cloud computing services”, rose $65.9 million: about 8 cents of each added revenue dollar. It was 11.0% of revenue in the quarter, against 12.4% a year earlier, by our arithmetic.
The fastest-growing cost line was research and development, at $101.8 million against $43.9 million, about 2.3 times. Sales and marketing was $68.8 million against $48.6 million, and general and administrative $64.5 million against $58.8 million. Income from operations was $1.71 billion, 78.2% of revenue, against 76.8% a year earlier, by our arithmetic.
AppLovin’s net margin from continuing operations and adjusted EBITDA margin by quarter, Q4 2023 to Q3 2026, as listed in its 3Q26 Financial Update published on 4 November 2026. Adjusted EBITDA is a non-GAAP measure.
Wall St Engine’s earnings card for AppLovin’s third quarter of 2026, posted on 4 November. It lists revenue of $2.18 billion against a $2.08 billion estimate, EPS of $4.30 against $4.04, adjusted EBITDA of $1.84 billion and fourth-quarter revenue guidance of $2.37 billion to $2.40 billion against a $2.25 billion estimate. The piece checks each figure against AppLovin’s filed release.
What AppLovin Says Its AI Does
The results release describes AppLovin as “a leading marketing platform” and reports figures only. The company’s second-quarter 10-Q, filed on 5 August, ties revenue to its ad model: “We are able to grow our revenue by improving our various technologies, including improvements to our Axon AI recommendation system.” It also says “our revenue has grown rapidly, in particular since the launch of Axon AI, our advertising recommendation system.”
The same filing names the cost side: “datacenter costs as we continue to launch enhancements to our Axon AI recommendation system”. On new markets, it says “since 2024 we have begun expanding our customer base to include web-based e-commerce advertisers”, and that the company has “begun to introduce tools that create AI-generated interactive ads and video creatives for advertisers”.
The Cash and the Buybacks
Operating cash flow was $1,428.8 million in the quarter and free cash flow $1,418.4 million, about 99% of net income, by our arithmetic. AppLovin says it “repurchased and withheld 4.1 million shares of our Class A common stock, for a total cost of $1.4 billion”, ending the quarter with 331 million shares outstanding.
The board raised the repurchase authorization by $3.2 billion, “bringing our total current share repurchase authorization to $3.6 billion as of the end of October”. Cash was $3.17 billion at 30 September and long-term debt $3.52 billion.
Twelve Quarters of Margin
The financial update posted with the results lists twelve quarters. Revenue went from $576.5 million in the fourth quarter of 2023 to $2,184.8 million, about 3.8 times, by our arithmetic. Net margin from continuing operations went from 33% to 66% over the same span, and the adjusted EBITDA margin from 71% to 84%.
The fourth-quarter revenue guidance is 43% to 44% above the $1,657.9 million reported for the fourth quarter of 2025, by our arithmetic. Wall St Engine’s earnings card put consensus at $2.08 billion of third-quarter revenue, $4.04 of EPS and $2.25 billion of fourth-quarter revenue.
What the Risk Factors Say
The third-quarter 10-Q had not been filed on EDGAR when we read the filings, so there is no new risk-factor text to compare yet. The second-quarter 10-Q carries this heading: “The development and use of AI in our business, combined with an uncertain regulatory environment, may adversely affect our business, reputation, financial condition or results of operations.”
As we read the release, the added costs went mostly to research and development and cost of revenue: research and development rose to $101.8 million from $43.9 million, and cost of revenue rose $65.9 million.
As we read the twelve-quarter table, the net margin from continuing operations doubled from 33% to 66% while revenue rose about 3.8 times, by our arithmetic.
AppLovin Form 10-Q, second quarter 2026
“We are able to grow our revenue by improving our various technologies, including improvements to our Axon AI recommendation system.”
Three Implications
FOR THE COST BASE Cost of revenue was 11.0% of revenue in the quarter, against 12.4% a year earlier, by our arithmetic; AppLovin does not report a gross margin.
FOR CASH Free cash flow was $1,418.4 million, about 99% of net income, by our arithmetic, and the company spent $1.4 billion repurchasing and withholding 4.1 million shares.
FOR THE GUIDE Fourth-quarter revenue guidance of $2,365 million to $2,395 million is 43% to 44% above the fourth quarter of 2025, by our arithmetic, with an adjusted EBITDA margin of 84% to 85%.
The framework says: “Inference is a revenue engine: continuous, distributed across millions of applications and agents, incurring marginal cost with every query.”
As we read it, AppLovin’s release puts a number on that marginal cost for one ad business: cost of revenue, mostly datacenter costs per its 10-Q, was 11.0% of revenue in the quarter, down from 12.4%, by our arithmetic. The framework does not discuss AppLovin.
The results release and financial update we read give no e-commerce revenue, advertiser count, referral or self-serve figure, and no split of revenue by product. We did not hear the 5:00 PM ET webinar. AppLovin does not report a gross margin: the cost-of-revenue share is our arithmetic, and that line also includes amortization and payment processing fees, not only datacenter costs. The 10-Q quotes come from the second-quarter filing, not from a third-quarter statement.
This Week In The AI Supercycle
Anthropic’s Reality Check, The Capital Squeeze, & The New World Order
This week’s issue: for two years the AI build-out was priced as a growth story. This week it started being priced as a credit story.
AppLovin added about $780 million of quarterly revenue from a year earlier and about $630 million of operating income, about 81 cents per added dollar, by our arithmetic, while cost of revenue, mostly datacenter costs per its 10-Q, fell to 11.0% of revenue from 12.4%. Its latest 10-Q says revenue has grown rapidly since the launch of Axon AI, its advertising recommendation system.
95,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.
A note on sourcing. Figures are from AppLovin’s third-quarter 2026 results release, filed with the SEC on 4 November 2026, and the 3Q26 financial update on its investor relations website. Quotations about Axon AI, datacenter costs and e-commerce are from its second-quarter 2026 Form 10-Q, filed on 5 August 2026. Consensus figures are as cited by Wall St Engine. Changes, ratios and multiples are our arithmetic.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
