Tinder says it no longer relies on a “desirability” score and that it now adjusts matches a user sees within 24 hours of someone acting on that user's profile
The Method Behind Our Matching Jack Morse / Mashable : Tinder just made it harder to know why you're seeing awful profiles Tweets: Steve Kovach / @stevekovach : This explains everything about my Tinder experience back in the day. Literally 0% success rate. http://twitter.com/... Ashley Carman / @ashleyrcarman : happy friday news dump : ) no idea why tinder published this at 530PM, BUT they finally gave some clarity around how their algorithm works and they talk about the death of the ~elo score~ http://www.theverge.com/...
Context & Ripple Effects
Tinder's matching has been an open question since it confirmed in 2016 that it calculated a desirability rating for every user from swipe data, a single internal number that shaped who saw whom. The company had already shown it was comfortable quietly reordering what users see when it launched [[a:876042|Smart Photos, which alternated profile photos based on interest and lifted matches up to 12% in testing]].
This disclosure walks back the score framing entirely: Tinder now says matches adjust within 24 hours of someone acting on your profile, replacing a static rating with a fast-moving feedback loop. The timing drew notice — Ashley Carman flagged the late-Friday news dump — because it lands amid growing pressure on dating platforms to explain why users see the profiles they do.
First-order effects
- Users get a materially faster feedback loop: acting on someone's profile reshapes the deck they're shown within a day, rather than being mediated by a cumulative desirability number.
Second-order effects
- With no single score to optimize, the levers that demonstrably move visibility — like Smart Photos-style photo testing — become the main thing users and Tinder itself can tune, pushing product effort toward surface-level signals over long-run ratings.
Third-order effects
- If the pattern holds, matching systems across dating apps converge on continuously updated behavioral signals instead of fixed rankings — a shift consistent with Tinder's later moves to rebuild the profile experience around prompts and quizzes and its pledge to prioritize user experience over monetization through 2026.
The trend: Dating platforms are trading opaque one-number rankings for fast-updating behavioral signals, with disclosure arriving under pressure to explain algorithmic matching to users.