Personality Compatibility: What the Science Really Says

The Myth of the Perfect Match

Open dating apps and you will find compatibility scores, personality-based matching algorithms, and quizzes promising to identify "your type." The pitch is seductive: answer forty questions and the algorithm will locate someone whose temperament fits yours like a key in a lock. It is also, according to the people who study long-term relationships for a living, mostly wishful thinking.

The idea that personality similarity, or its opposite, complementarity, drives relationship success has been tested for decades on tens of thousands of couples, and the results keep landing in the same disappointing place: personality match explains very little about who stays together and who does not.

What Similarity Research Actually Shows

The similarity-attraction hypothesis dates back to psychologist Donn Byrne's work in the 1960s and 1970s, which found that strangers rated hypothetical partners more favorably when their attitudes lined up. That finding held up reasonably well for first impressions. It did not hold up nearly as well for marriages.

In 2010, a team led by Portia Dyrenforth analyzed personality and relationship satisfaction data from more than 15,000 couples across the United States, United Kingdom, Germany, and Australia. Their conclusion was blunt: actual similarity between partners' Big Five traits accounted for less than half a percent of the variance in relationship satisfaction. When they simulated random pairings of the same people, the resulting compatibility scores looked almost identical to those of the real couples. Swap partners at random within the sample, in other words, and the satisfaction predictions barely moved.

What did matter, consistently, was each person's own trait level, independent of their partner's. Someone low in neuroticism tends to report a happier relationship whether their partner is calm or anxious, because they bring less reactivity and fewer negative interpretations to ordinary friction.

Complementarity: The Weaker Cousin

The competing folk theory, that opposites attract and balance each other out, traces to psychologist Robert Winch's complementary needs theory from the 1950s. Winch proposed that a dominant person would be happiest with a submissive one, an emotionally expressive partner with a reserved one, and so on down the list of traits. It is an appealing story, and it has been tested repeatedly since.

Support for the idea has never been strong. A handful of narrow studies find mild complementarity effects in specific domains, dominance and submission in conversational dynamics being the clearest example, but broad personality complementarity across the Big Five shows no consistent link to satisfaction once individual trait levels are controlled for. Most reviews of the literature conclude the effect, if it exists at all, is too small to build a matching product on.

Part of the appeal of complementarity theory is that it explains specific couples people already know, the quiet accountant married to the gregarious salesperson, and treats the pairing as proof of a rule. That reasoning runs backward. Researchers can find a complementary-seeming couple in almost any sample and a similar-seeming couple right next to them reporting the same satisfaction score, which is exactly what you would expect if the trait pairing itself were doing very little of the work.

What Actually Predicts Relationship Success

If matching traits is not the answer, what is? Individual differences still matter, just not in the way dating apps market them. Low neuroticism in either partner is one of the most consistent predictors of relationship satisfaction found across the literature. High conscientiousness helps too, mostly because conscientious people follow through on commitments and handle conflict without letting it escalate.

Behavior matters more than trait profile. John Gottman's decades of observational research on married couples identified specific interaction patterns, criticism, contempt, defensiveness, and stonewalling, that predict divorce with striking accuracy, sometimes years ahead of the split. None of those patterns maps cleanly onto any personality type. A calm, agreeable person can still stonewall under enough stress, and a highly neurotic person can learn to repair conflict well.

Perceived partner responsiveness, whether you believe your partner understands, validates, and supports you, shows up again and again as a stronger predictor of satisfaction than any trait comparison. Shared values around money, family, and how to spend time tend to matter more day to day than whether both partners score high on openness.

Where Personality Tests Still Help

None of this means personality tests are useless for couples. Understanding your own patterns, and naming them out loud to a partner, gives both people a vocabulary for what would otherwise look like unexplained friction. Taking a Big Five assessment and comparing notes can turn "you never want to talk about this" into "I process stress by needing quiet first, and you process it by needing to talk it through right away."

The mistake is treating the results as a verdict on compatibility rather than a starting point for conversation. A couple who wants a clearer read on how each of them handles stress, conflict, and decisions under pressure can get a verified profile and use it to plan around real differences, instead of hoping a matching score will do the relationship work for them.

It also helps to remember what a compatibility percentage cannot tell you. It cannot tell you whether two people share the same appetite for honesty during a disagreement, whether they agree on how much time apart feels normal, or whether either of them is willing to repair things after a bad week. Those questions get answered through actual conversation and shared history, not through a shared four-letter code or a matching trait score generated before the first date.