Match Group’s product development expense rose 14% to $119.8 million in the third quarter, which its 10-Q attributes primarily to “utilization of tokens associated with AI” at Tinder and Hinge and to more staff at Hinge. It was the largest year-on-year rise of the three quarters of 2026, by our arithmetic from its filings.
Match saved more on app stores than it added. Payers moving to alternative payment methods cut Match’s in-app purchase fees by $43.0 million, the 10-Q says. The $14.8 million rise in product development equals 34% of that saving, by our arithmetic: about 34 cents of each dollar saved on store fees.
Revenue fell 3% to $891 million. Tinder’s daily active users returned to year-on-year growth in September for the first time in three and a half years, according to the company’s prepared remarks.
Business Pill · Tokens
An evergreen Business Engineer explainer on tokens: the unit AI models read, write and bill in. It explains the general idea only and does not discuss Match Group.
The key insight: As we read the filings, AI has become a named operating cost at Match: token use is the first driver the 10-Q gives for a 14% rise in product development, in the same quarter that lower app-store fees cut cost of revenue by 16%.
Product development was 13% of revenue, against 11% a year earlier, the 10-Q says. Match does not give a separate figure for token costs. Over nine months, product development rose 3%, with token use and Hinge headcount partly offset by lower stock-based compensation and more capitalised salaries.
The quarterly filings show the turn. Product development was 3.3% lower year on year in the first quarter and 0.3% higher in the second, before the 14% rise in the third, by our arithmetic from the 10-Q figures.
The prepared remarks describe two uses. Match says it has “fundamentally changed our approach to recommendations by using AI and machine learning”, starting with a unified recommendation system called Queue Unification, and plans to begin testing real-time adaptive recommendations in the fourth quarter.
It also cites Missed Connections, a feature tested in Canada and Australia that surfaces people users may have crossed paths with. Match says the feature “went from idea to final product in just a few weeks, which in years past would have taken months or quarters”, and that in tests women were about four times as likely to like these profiles.
Match Group operating cost lines as a share of total revenue, third quarter of 2025 against third quarter of 2026, from the statement of operations in its Q3 2026 prepared remarks. Shares of revenue are our arithmetic; the drivers are as stated by Match in its prepared remarks and 10-Q.Animated: Match Group’s product development expense as a share of total revenue, third quarter of 2025 to third quarter of 2026: 11.5%, 12.4%, 13.5%, 13.5% and 13.4%. The shares, and the fourth-quarter 2025 figure (full year minus nine months), are our arithmetic from its 10-Q and 10-K filings.
Wall St Engine (@wallstengine) posted Match Group’s third-quarter highlights at 21:15 UTC on 2 November. Per the post: revenue of $891.2 million against an estimate of $892 million, EPS of $0.73 against $0.72, Adjusted EBITDA of $344 million, 13.8 million payers, Tinder revenue of $480 million and Hinge revenue of $221 million. The piece checks each figure against Match’s earnings release; the estimates are the post’s own, with no named source.
Where the Savings Came From
Cost of revenue fell 16% to $206.5 million, or 23% of revenue from 27%. The 10-Q attributes the fall to payers shifting from app store payments to alternate payment methods: $43.0 million lower in-app purchase fees, partly offset by $4.5 million more in credit card processing fees.
Selling and marketing rose $16 million, or 9%, on more spending at Tinder and Hinge. General and administrative costs fell 38%, because the prior-year quarter carried a legal settlement charge at Tinder. Adjusted EBITDA rose 14% to $344 million; excluding that $61 million charge, it fell 5%, the remarks say.
Tinder, Hinge and the Rest
Tinder’s direct revenue fell 2% to $480 million. Payers fell 4% to 8.9 million, while revenue per payer rose 2% to $18.02. Daily active users were down 1% for the quarter, monthly active users 5.5%. The remarks put the hit from user-experience tests and product changes at about $5 million.
Hinge’s direct revenue rose 19% to $221 million, with payers up 23% to 2.3 million and revenue per payer down 3% to $31.81 after discounts and a lower-priced tier. Everyone Everywhere, the segment that combines Match’s former Evergreen and Emerging and MG Asia segments, fell 20% to $176 million. Group payers fell 5% to 13.8 million.
Guidance and Cash Returns
Match guided fourth-quarter revenue to $865 million to $875 million and Adjusted EBITDA to $327.5 million to $332.5 million. It now expects a full-year Adjusted EBITDA margin of 38.8%, against the 37.5% it guided in February, citing “additional alternative payment savings, as well as lower compensation expense and operating costs”.
Through September, Match generated $832 million of free cash flow, bought back 9.6 million shares for $336 million at an average of $35 and paid $137 million in dividends.
How the Figures Compare With Estimates
Wall St Engine, posting on X at 21:15 UTC on 2 November, put estimates at $892 million for revenue, $0.72 for earnings per share and $332 million for Adjusted EBITDA, and $871 million for fourth-quarter revenue. It did not name the source. Revenue of $891.2 million came in slightly below that figure; earnings per share of $0.73 and Adjusted EBITDA came in above.
What the Risk Factors Say
The 10-Q’s risk-factor item lists no new factors and refers readers to the annual report for 2025. Its cautionary list of risks includes “challenges with properly managing the use of artificial intelligence”.
As we read the prepared remarks, Match describes AI in two places: recommendations that it says optimise for Sparks, retention and ecosystem health, and the speed of building features such as Missed Connections.
As we read the cost lines, spending moved from fees paid to app stores toward Match’s own product and marketing: cost of revenue fell to 23% of revenue, while product development rose to 13% and selling and marketing to 21%.
As we read the guidance, the payment savings are still carrying the margin: Match raised its full-year Adjusted EBITDA margin outlook to 38.8% from 37.5%, citing additional alternative payment savings.
Match Group, Q3 2026 Prepared Remarks, 2 November 2026
“This feature is a clear example of how we’re using AI to accelerate the pace of productinnovation”
Three Implications
FOR TINDER Daily active users returned to year-on-year growth in September, Match says, while payers fell 4% and direct revenue 2% in the quarter.
FOR APP STORES Lower in-app purchase fees saved Match $43.0 million in the quarter as payers moved to alternative payment methods, partly offset by $4.5 million more in card fees, the 10-Q says.
FOR AI MODEL PROVIDERS Match names token use at Tinder and Hinge as a driver of its product spending but does not disclose the amount or the providers it uses.
The framework opens its argument this way: “For two years, AI was a CIO line item and a CEO talking point. In 2026 it became a CFO problem, because the economics finally turned legible — and unforgiving.”
As we read it, Match’s 10-Q is a small example of that shift: token use now appears as a stated driver of an operating expense line. The framework, published in June 2026, does not discuss Match Group.
What Is Not Established
Match does not say how much it spent on AI tokens, which models or providers it uses, or how token costs split between Tinder and Hinge. The 34% figure compares two separate line items; Match does not link the fee savings to its AI spending. We did not review the earnings call.
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.
Match’s third quarter put AI tokens on its cost lines: product development rose $14.8 million, which the 10-Q ties primarily to token use and Hinge staff, while lower app-store fees saved $43.0 million. Tinder’s daily users turned up in September, while group revenue fell 3%.
95,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.
A note on sourcing. Company figures come from Match Group’s Q3 2026 earnings release and prepared remarks (8-K Exhibits 99.1 and 99.2, filed 2 November 2026), its 10-Q for the quarter, and quarterly figures from its 10-Q and 10-K filings via the SEC’s XBRL data. Estimates are Wall St Engine’s. The 34% figure (the $14.8 million product development rise divided by the $43.0 million fall in in-app purchase fees), the quarterly growth rates and the shares of revenue in the chart are our arithmetic. We did not contact Match Group.
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.
