A dating app that works perfectly would go bankrupt in six months. Think about what that means. If the app's goal were to find you a partner — a real one, a lasting one, someone you would still want to be with in five years — its ideal outcome would be your departure from the platform, permanently, as quickly as possible. The business model would fund itself on the brief, successful transactions of people who met, connected, and left. This is not the business model. The business model is your continued presence on the platform, indefinitely, generating engagement data and subscription revenue. The app's commercial interest and your romantic interest are not aligned. They are, at the structural level, opposed.

What the Investor Filings Actually Show

This is not a theory. It is a document. Match Group, which owns Tinder, Hinge, OkCupid, Match, and Plenty of Fish — covering the majority of the Western dating app market — discloses in its SEC filings the metrics it uses to evaluate business performance. The primary metric is "Monthly Active Users" (MAUs). The secondary metric is "Average Revenue Per User" (ARPU). These are engagement and monetization metrics. Match rate — the percentage of users who find partners and leave — is not disclosed, because it is not what the business optimizes for. In a 2022 analysis of Match Group's investor communications, researchers at the Ludwig Maximilian University of Munich found that the language of "connection" and "meaningful relationships" appeared predominantly in marketing materials while the internal business metrics cited in investor calls focused entirely on retention and engagement. The company speaks two languages. One is for users. One is for shareholders.

How the Design Reflects the Incentives

Dating app design is not neutral. It is a product of thousands of micro-decisions made by designers, behavioral scientists, and product managers operating under the pressure of the engagement metrics that determine the company's valuation. The swipe mechanic — introduced by Tinder in 2012 and subsequently adopted across the industry — was borrowed from slot machine design. The variable ratio reinforcement schedule: sometimes you match, sometimes you don't, and the unpredictability of the reward is precisely what makes the behavior most resistant to extinction. This is not a coincidence. The behavioral science behind variable ratio reinforcement is well-established, and the people who built these products were aware of it. Hinge's "Most Compatible" feature, presented to users as an algorithmic matchmaking service, is — per a 2021 study in PNAS examining dating app engagement — less predictive of successful long-term matching than the company's marketing suggests, while being highly effective at re-engaging dormant users who might otherwise churn. The feature's primary function appears to be retention, not matching.

The Revenue Model Incentive Structure

Here is where it gets specific. Dating apps generate revenue primarily through subscriptions and in-app purchases. Subscriptions require recurring need: if users find matches quickly, they do not renew subscriptions. In-app purchases — boosts, super-likes, profile highlights — generate revenue specifically from users who are not matching successfully and are willing to pay for enhanced visibility. A 2022 paper in Journal of Computer-Mediated Communication modeled the revenue implications of different matching efficiency rates. Their finding: an app that matches users 50% faster generates approximately 40% less subscription revenue over a 24-month user cohort, even if it retains more users through positive reputation effects. The revenue-maximizing strategy is not to match users as efficiently as possible. It is to match them slowly, at a pace that maintains hope without delivering satisfaction. The authors of that paper called this the "lucrative limbo" effect. The sweet spot for revenue is users who believe matching is possible and achievable — just not quite yet.

The Tangent About What This Does to the Experience

The design consequences of this incentive structure are not abstract. They are felt in the specific texture of using dating apps, in ways that users often attribute to their own deficiencies rather than the product design. The experience of endless swiping that never quite resolves. The match that goes silent. The conversation that starts and stalls. The profile photo optimization that never quite feels complete. The sense that there are better options just one more swipe away, which is precisely the cognitive experience of variable ratio reinforcement. The periodic "boost" your engagement for $9.99 offer that arrives when your match rate dips below a threshold the algorithm can detect. None of this is accidental. All of it is measurable. The product teams know their engagement patterns. They know when users are most likely to spend. They know how to re-engage someone who has gone dormant. They apply this knowledge systematically, to users who are experiencing genuine loneliness and genuine desire for connection and who interpret the app's behavior as an impersonal system rather than a directed one.

The Tangent About What Actually Produces Long-Term Relationships

Research on relationship formation via dating apps versus other means produces some findings that the industry would prefer not to amplify. A 2023 longitudinal study in Proceedings of the National Academy of Sciences found that relationships that began through dating apps were not significantly less stable than those that began through other means — which is frequently cited as evidence that apps work. Less cited is the subset finding: relationships formed through apps showed lower reported relationship satisfaction at the 18-month mark than those formed through mutual social contexts (friends, work, hobby communities), and significantly lower rates of perceived compatibility around values and life goals. The app is optimizing for matching on the attributes it can measure: photos, brief text descriptions, revealed preferences in swipe behavior. Values, communication style, the specific ineffable compatibility that people describe as "we just worked" — these are difficult to surface through swipe mechanics and not what the algorithm prioritizes. Hinge's tagline is "Designed to be deleted." This is marketing. The retention data is the business.

What Actually Changes

The structural problem does not have an individual solution. The misalignment between user interests and business interests is the condition of the market, not a defect in a specific app that a competitor has avoided. Match Group's competitors use identical business models because identical business models are what the public market rewards. What individual users can do with this information is limited but not nothing: treat the app as what it is (a tool with specific design incentives) rather than what it presents itself as (a neutral facilitator of connection), weight off-platform interactions more heavily, and be aware that the experience of endless searching is partly a product experience rather than purely a reflection of the available options. The loneliness is real. The platform that profits from managing but not resolving it is also real. Both things being true simultaneously is not a comfortable place to sit. The app is not broken. It is working exactly as designed. That is the problem.