OKCupid weighed in on the controversy around Facebook’s experiments on Monday by admitting that it, too, has lied to users in the past.
Only there’s one huge difference: Unlike Facebook, which admitted to purposefully manipulating the emotions of its users, OKCupid said it lies to users to help improve their matchmaking system and help users find love.
Christian Rudder, one of the founders of OKCupid, wrote an enticing blog post called “We Experiment On Human Beings!” to lay out three different experiments the dating site ran on its users without asking for permission.
“Guess what, everybody: if you use the Internet, you’re the subject of hundreds of experiments at any given time, on every site,” Rudder said. “That’s how websites work.”
Rudder starts off the post with the following disclaimer:
I’m the first to admit it: we might be popular, we might create a lot of great relationships, we might blah blah blah. But OkCupid doesn’t really know what it’s doing. Neither does any other website… Experiments are how you sort all this out.
He goes on to explain three experiments OKCupid has performed on its users, which allegedly helped the site do a better job at matching its users.
The first experiment is the most benign since it was entirely public. Last year, OKCupid proclaimed “Love Is Blind Day” on Jan. 15 and turned into a blind date app, removing all photos of its users. In those seven photo-less hours, OKCupid noticed new conversations dipped dramatically, but people responded to first messages 44% more frequently and contact details were exchanged more quickly.
“In short, OKCupid worked better,” Rudder explained.
The second experiment helped OKCupid determine the best way to feature profiles. In an AB test, some users saw profiles with just photos and others saw both photos and text. And the company discovered people care about photos much more so than words. It wasn’t too shocking, but the information was useful nonetheless for the site hoping to match up its users.
The third experiment may have been the most controversial, in that OKCupid actually lied to users about what they were seeing on the website. The company took pairs of users with low match ratings of around 30% (the ideal match is 100%) and told them they were a 90% match. They also did the opposite, giving highly-compatible pairs of users low match ratings.
OKCupid found, somewhat unsurprisingly, that more people sent messages to matches they perceived to be compatible with. But users also sent more messages to people they were actually compatible with, regardless of the match score shown to them, proving the efficacy of its matching algorithm, but also the role of suggestion in matchmaking.
So fine, OKCupid didn’t warn users it was messing with them, but these kinds of experimentations can actually help the dating site do a better job at finding you a date, so is that really so bad?