About 3 in 10 American adults have used a dating site or app at some point, according to Pew Research Center’s 2023 data. That number sounds like a solid foundation for modern romance until you look at the other side of it. Only 1 in 10 partnered adults say they met their current partner through one of these platforms. The gap between adoption and outcome is wide, and it raises a fair question about what these matching systems are actually doing for the people who use them.
Most platforms promise some version of the same thing: feed in your preferences, answer a few questions, and let the algorithm find someone compatible. The pitch is appealing because it sounds rational. But the results suggest that the connection between algorithmic matching and lasting relationships is weaker than most users assume. A growing body of research points to reasons why, and some of those reasons have less to do with the technology and more to do with the business model behind it.
The Placebo Problem With Compatibility Matching
Lara Sharabi, a communications researcher, published a study in 2021 that found something worth paying attention to. People who believed an algorithm had matched them well reported better first dates. But the quality of those dates had little to do with how well the algorithm actually performed. The effect was closer to a placebo, where positive expectations produced positive outcomes regardless of the system’s accuracy. Sharabi’s findings, discussed in the Harvard Data Science Review, suggest that what people feel about being matched matters more than the match itself.
This is a problem for anyone relying on a platform to screen for long-term compatibility. If the algorithm’s main contribution is making you optimistic before a date, the matching system itself is not doing the heavy lifting. Your own hopefulness is. The technology provides a framework that feels scientific, and that feeling of precision gives users confidence. But confidence before a first date and compatibility over years of partnership are 2 very different things.
What People Actually Want Varies More Than Any Algorithm Can Sort
Dating apps assume that compatibility can be reduced to a set of measurable inputs. But people look for very different things depending on where they are in life. Some users want casual dates, others are focused on meeting rich guys, and plenty are searching for someone to build a long-term partnership with. A single matching system cannot meaningfully serve all of these goals at once, and most platforms do not try to separate them in any honest way.
This mismatch between user intent and platform design creates friction that no algorithm can resolve. When everyone is funneled through the same swiping mechanism, the system rewards surface-level appeal over actual alignment in values or life direction. The Carnegie Mellon Tepper School study confirmed this, finding that platforms favor attractive profiles through a built-in popularity bias rather than sorting for genuine compatibility. The tool, in other words, is not broken. It was built for engagement, not for partnership.
Engagement Is the Product, Not Your Relationship
A study from Carnegie Mellon’s Tepper School of Business made a point that often gets overlooked in conversations about dating apps. The researchers found that platforms may prioritize keeping users engaged on the app over helping them find a lasting partner. That finding reframes the entire premise. If the business model depends on users continuing to swipe, there is a financial incentive to keep people searching rather than succeeding.
This does not mean companies are deliberately sabotaging your love life. But it does mean the algorithm’s goals and your goals are not perfectly aligned. A user who finds a committed partner in the first month and deletes the app is a lost subscriber. A user who stays active for 8 months, upgrading to premium features along the way, is far more valuable. The system does not need to be malicious to produce poor outcomes for people looking for permanence. It only needs to follow its own logic.
What the Mental Health Data Shows
A 2025 meta-analysis covering 23 studies and more than 26,000 participants found that dating app users reported worse psychological health and well-being compared to people who did not use the apps. That is a large sample size, and the consistency across studies makes it hard to dismiss. The research was published in Computers in Human Behavior.
The reasons behind this are likely tangled. Constant rejection, the volume of superficial interactions, and the addictive design of swipe-based interfaces all play a role. For someone who has been on a platform for months without a promising connection, the toll adds up. And the worse someone feels, the less likely they are to present their best self on a date, which creates a cycle that the algorithm has no mechanism to address.
Regulatory Actions Tell Their Own Story
The business practices of major dating platforms have drawn attention from regulators. In August 2025, Match Group agreed to a $14 million settlement over allegations of deceptive subscription practices, as reported by TechCrunch. Then in March 2026, the FTC alleged that OkCupid had shared nearly 3 million user photos with a facial-recognition company called Clarifai without user consent. That case resulted in a settlement that bars future privacy misrepresentations.
These cases do not directly relate to algorithmic matching, but they say something about how these companies treat their users. When a platform is willing to share your photos with a third party without telling you, the relationship between company and user is transactional in a way that should give people pause.
So Where Does That Leave You?
A 2024 SSRS poll found that 42% of people who tried online dating entered a committed relationship with someone they met on a platform. That is not a small number, and it would be inaccurate to say the apps never work. People do meet partners through them. But the data also suggests that the algorithm itself deserves less credit than the platforms claim. Your own effort, your willingness to show up honestly, and plain luck seem to matter at least as much as any matching formula. Treating the app as one tool among many, rather than a reliable system, is probably the most reasonable approach.
