The conflict of interest nobody talks about
Here is a question that should make you uncomfortable: what happens to a dating app's revenue when it actually works?
If you find a meaningful connection in your first month, you cancel your subscription. You delete the app. You tell your friends it worked. And the company loses a paying customer.
Now consider the alternative. You do not find anyone. You keep swiping. You get frustrated. You see a prompt offering you more visibility, more likes, a "boost" that puts you at the top of someone's stack. You pay. You keep paying. You stay.
This is the fundamental contradiction at the heart of modern dating apps. The product's success - helping you find someone - is the business model's failure. Every meaningful connection is a churned subscriber. Every lasting relationship is lost revenue.
Researchers have given this a name: the conflict of interest theory. A 2024 study analysed over 7,000 Tinder reviews and found that users widely believe dating apps are manipulating their visibility, suppressing good matches, and deliberately providing poor-fit suggestions to keep them engaged longer. The researchers noted the idea was so familiar to academics in the space that it barely needed explaining.
A class-action lawsuit filed against Match Group in 2024 put it more bluntly: the platforms are engineered to lock users into a "perpetual pay-to-play loop" at the expense of their actual relationship goals.
Match Group: the company behind (almost) everything
Most people do not realise that a single corporation - Match Group - owns Tinder, Hinge, OkCupid, Match.com, Plenty of Fish, The League, and dozens of other dating platforms. If you have used a dating app in the last decade, there is a strong chance Match Group made it.
This is not inherently a problem. Large companies can build good products. But Match Group's track record raises serious questions about whose interests the platforms are designed to serve.
💰 $14 million FTC settlement (2025). Match Group agreed to pay $14 million and overhaul its cancellation and billing practices after the FTC found the company had used deceptive advertising, made cancellation unnecessarily difficult, and retaliated against users who initiated chargebacks. The "six-month guarantee" that promised a free renewal if you had not met someone? It came with hidden conditions that were nearly impossible to satisfy.
⚠️ Safety failures. A 2025 investigation by NPR and The Markup found that users accused of sexual assault were able to continue using Match Group apps for years after being reported. In one case, a convicted predator was actually promoted as a "standout" date on Hinge - after survivors had already reported him in the app.
🔒 Data retention concerns. Reports indicate Match Group may retain and use personal data even after banning users - including from accounts that can no longer log in. Transparency around bans, appeals, and data handling remains minimal.
When a single company controls the majority of the market, there is limited competitive pressure to fix these problems. If you are banned from one Match Group app, you may lose access to all of them - with no transparent appeals process. If you are frustrated with Tinder, the next app you try is probably also theirs.
Dark patterns by design
The conflict of interest is not theoretical. It shows up in the product design itself.
- Artificial scarcity. These mechanics are part of a broader dopamine loop designed to keep you engaged. Free users get a limited number of daily likes. Run out? Pay for more, or wait until tomorrow. The limit is not about quality - it is about creating urgency.
- Hidden visibility. Your profile's reach is quietly throttled. You see "likes" from blurred-out profiles - pay to reveal them. The app knows who is interested in you and deliberately hides it behind a paywall.
- Boost mechanics. Pay to temporarily become more visible. The implication: you are less visible by default. The platform controls who sees you, and charges you to be seen.
- Opaque algorithms. No dating app tells you why someone was suggested. The matching logic is a black box. You cannot see the criteria, challenge the results, or understand what the app thinks "compatible" means.
- Shadowbanning. Tinder has never officially confirmed it, but users have documented it extensively: delete and recreate your account too often, and your match rate drops to near zero. The platform silently punishes behaviour that avoids paying for a "reset" - which they sell as a premium feature.
- Cancellation friction. The FTC found that Match Group made it deliberately difficult to cancel subscriptions. This is not accidental poor UX - it is a business decision.
These are not edge cases. They are core mechanics of the most popular dating apps in the world. Every one of them serves the same purpose: keeping you engaged and paying, regardless of whether you are actually finding meaningful connections.
Not everything is broken
It would be dishonest to say every dating app is equally bad. Some platforms have implemented genuinely good ideas - even if they fall short in other areas. Understanding what works is just as important as understanding what does not.
✓ Massive user base. Low barrier to entry. Culturally ubiquitous. If you want volume and speed, nothing else comes close.
✗ But volume is not depth. Tinder is optimised for quick decisions based on photos - not for facilitating the kind of deep, meaningful connections that actually lead to lasting relationships. The swipe mechanic encourages snap judgements. Conversations frequently go nowhere because two people who matched on appearance alone discover they have nothing in common. Owned by Match Group with all the dark patterns listed above.
✓ The best data-matching approach I have seen on a major platform. OkCupid's Q&A system is genuinely smart - users answer questions about their values, lifestyle, and preferences, then indicate which answers they would accept from a match and how important each question is. This creates a structured compatibility dataset that goes far beyond photos and bios. The transparent compatibility percentage is a real differentiator.
✗ The Q&A is self-reported - no behavioural data from external platforms. The user base has declined significantly. The UX feels dated compared to competitors. And it is owned by Match Group, which means the same corporate incentives and dark patterns apply. Good data model, questionable stewardship.
✓ "Designed to be deleted" is a strong brand promise. The prompts encourage personality. The UX is polished. Hinge genuinely tries to position itself as the anti-Tinder.
✗ Still photo-first. The algorithm is just as opaque as Tinder's. Premium pricing has crept above £50/month. And the brand promise has not stopped Match Group (who own Hinge) from implementing the same paywall mechanics: hidden likes, visibility boosts, limited daily interactions unless you pay. The ethos is right; the execution is still constrained by the parent company's revenue model.
✓ The women-message-first mechanic was a genuine innovation when it launched. The expansion into BFF and Bizz shows ambition beyond dating.
✗ The 24-hour match expiry creates artificial pressure. Premium pricing is aggressive. The core differentiation - who sends the first message - does not change how matches are made in the first place. Bumble's matching is still photo-and-bio-driven with an opaque algorithm. The differentiator is messaging etiquette, not compatibility.
✓ A dating app built specifically for music lovers, with over 2 million users. You fill in your favourite artists, songs, and genres, create blind tests, and the algorithm matches based on musical taste. The concept is sound: your music taste is a genuine identity signal, and building a dating experience around it makes more sense than a generic bio.
✗ Limited to a single interest category. If music is your primary identity, it works - but most people's compatibility extends across multiple dimensions. The user base can be thin outside major cities. It demonstrates that niche, interest-driven matching resonates with users - the challenge is scaling that approach across more of who you are.
The pattern
Look at the landscape honestly and a pattern emerges:
- The platforms with the best reach (Tinder, Bumble) have the shallowest matching.
- The platform with the best data model (OkCupid) is declining under corporate ownership.
- The platforms with the best intentions (Hinge) are still constrained by the same revenue model.
- The niche platforms (Turn Up, and others) prove the concept works - but only for a single dimension of who you are.
Nobody is doing all of it. Nobody is combining real behavioural data across multiple interest categories, weighting niche overlap more than mainstream overlap, making the algorithm transparent, and building the whole thing without a business model that profits from keeping you searching.
What this means for Affinity Atlas
This is the landscape Affinity Atlas exists in. Not because the world needs another dating app - but because it needs a different kind of dating app.
- No conflict of interest. Affinity Atlas is an independent passion project. There are no investors demanding subscriber growth. There is no structural incentive to keep you swiping instead of connecting.
- Transparent matching. Every match comes with an explanation. You can see exactly which signals contributed, how they were weighted, and why someone was suggested. No black box.
- Behavioural data, not just self-reported. OkCupid's Q&A model is good. Affinity Atlas takes it further by combining Q&A with opt-in data from platforms you already use - Spotify, Steam, Untappd, Letterboxd, Strava, Goodreads, GitHub, and more. Your real-world passions become compatibility signals.
- Niche weighting. Sharing a love for an artist with a popularity score of 12/100 means more than both knowing whoever is topping the charts. The same principle applies across all categories - games, beer styles, film directors, running routes, programming languages.
- Multi-dimensional. Unlike single-category niche apps, Affinity Atlas matches across all of your interests simultaneously. Music, gaming, code, drinks, film, fitness, books, events - each weighted by how much it matters to you.
- Privacy by design. Every integration is opt-in, permission-scoped, and revocable. Raw data stays private. Only derived signals are used in matching and shown to other users.
If you are tired of the current state of dating apps - or if you are curious about what a transparent, data-respectful alternative looks like - try the demo or get in touch.
A different approach to dating
Transparent matching, real behavioural data, no conflict of interest. See how interest-based compatibility works.
Try the demo