Share All sharing options for: The Tinder algorithm, explained
This story is part of a group of stories called
If there’s one thing I know about love, it’s that people who don’t find it have shorter life spans on average. Which means learning how the Tinder algorithm works is a matter of life and death, extrapolating slightly.
According to the Pew Research Center, a majority of Americans now consider dating apps a good way to meet someone; the previous stigma is gone. But in , at the time of Pew’s survey, only 15 percent of American adults had actually used a dating app, which means acceptance of the tech and willingness to use the tech are disparate issues. On top of that, only 5 percent of people in marriages or committed relationships said their relationships began in an app. Which raises the question: Globally, more than 57 million people use Tinder – the biggest dating app – but do they know what they’re doing?
They do not have to answer, as we’re all doing our best. But if some information about how the Tinder algorithm works and what anyone of us can do to find love within its confines is helpful to them, then so be it.
The first step is to understand that Tinder is sorting its users with a fairly simple algorithm that can’t consider very many factors beyond appearance and location. The second step is to understand that this doesn’t mean that you’re doomed, as years of scientific research have confirmed attraction and romance as unchanging facts of human brain chemistry. The third is to take my advice, which is to listen to biological anthropologist Helen Fisher and never pursue more than nine dating app profiles at once. Here we go.
The Tinder algorithm basics
A few years ago, Tinder let Fast Company reporter Austin Carr look at his “secret internal Tinder rating,” and vaguely explained to him how the system worked. Essentially, the app used fastflirting tips an Elo rating system, which is the same method used to calculate the skill levels of chess players: You rose in the ranks based on how many people swiped right on (“liked”) you, but that was weighted based on who the swiper was. The more right swipes that person had, the more their right swipe on you meant for your score.
Tinder would then serve people with similar scores to each other more often, assuming that people whom the crowd had similar opinions of would be in approximately the same tier of what they called “desirability.” (Tinder hasn’t revealed the intricacies of its points system, but in chess, a newbie usually has a score of around 800 and a top-tier expert has anything from 2,400 up.) (Also, Tinder declined to comment for this story.)
In , Tinder published a blog post explaining that this Elo score was “old news” and outdated, paling in comparison to its new “cutting-edge technology.” What that technology is exactly is explained only in broad terms, but it sounds like the Elo score evolved once Tinder had enough users with enough user history to predict who would like whom, based solely on the ways users select many of the same profiles as other users who are similar to them, and the way one user’s behavior can predict another’s, without ranking people in an explicitly competitive way. (This is very similar to the process Hinge uses, explained further down, and maybe not a coincidence that Tinder’s parent company, Match, acquired Hinge in .)
But it’s hard to deny that the process still depends a lot on physical appearance. The app is constantly updated to allow people to put more photos on their profile, and to make photos display larger in the interface, and there is no real incentive to add much personal information. Most users keep bios brief, and some take advantage of Spotify and Instagram integrations that let them add more context without actually putting in any additional information themselves.