Who invented okcupid

McKinlay watched with satisfaction as his bots purred along. Then, after about a thousand profiles were collected, he hit his first roadblock. OkCupid has a system in place to prevent exactly this kind of data harvesting: It can spot rapid-fire use easily.

One by one, his bots started getting banned. He turned to his friend Sam Torrisi, a neuroscientist who'd recently taught McKinlay music theory in exchange for advanced math lessons. Torrisi was also on OkCupid, and he agreed to install spyware on his computer to monitor his use of the site. With the data in hand, McKinlay programmed his bots to simulate Torrisi's click-rates and typing speed. He brought in a second computer from home and plugged it into the math department's broadband line so it could run uninterrupted 24 hours a day.

After three weeks he'd harvested 6 million questions and answers from 20, women all over the country. McKinlay's dissertation was relegated to a side project as he dove into the data. He was already sleeping in his cubicle most nights. Now he gave up his apartment entirely and moved into the dingy beige cell, laying a thin mattress across his desk when it was time to sleep. For McKinlay's plan to work, he'd have to find a pattern in the survey data—a way to roughly group the women according to their similarities.

The breakthrough came when he coded up a modified Bell Labs algorithm called K-Modes. First used in to analyze diseased soybean crops, it takes categorical data and clumps it like the colored wax swimming in a Lava Lamp.

With some fine-tuning he could adjust the viscosity of the results, thinning it into a slick or coagulating it into a single, solid glob. He played with the dial and found a natural resting point where the 20, women clumped into seven statistically distinct clusters based on their questions and answers. He retasked his bots to gather another sample: 5, women in Los Angeles and San Francisco who'd logged on to OkCupid in the past month. Another pass through K-Modes confirmed that they clustered in a similar way.

His statistical sampling had worked. Now he just had to decide which cluster best suited him. He checked out some profiles from each. One cluster was too young, two were too old, another was too Christian.

But he lingered over a cluster dominated by women in their mid-twenties who looked like indie types, musicians and artists. This was the golden cluster. The haystack in which he'd find his needle. Somewhere within, he'd find true love. Actually, a neighboring cluster looked pretty cool too—slightly older women who held professional creative jobs, like editors and designers.

He decided to go for both. He'd set up two profiles and optimize one for the A group and one for the B group. He text-mined the two clusters to learn what interested them; teaching turned out to be a popular topic, so he wrote a bio that emphasized his work as a math professor.

The important part, though, would be the survey. He picked out the questions that were most popular with both clusters. He'd already decided he would fill out his answers honestly—he didn't want to build his future relationship on a foundation of computer-generated lies.

But he'd let his computer figure out how much importance to assign each question, using a machine-learning algorithm called adaptive boosting to derive the best weightings. With that, he created two profiles, one with a photo of him rock climbing and the other of him playing guitar at a music gig. Sex or love? Answer: Love, obviously. But for the younger A cluster, he followed his computer's direction and rated the question "very important.

When the last question was answered and ranked, he ran a search on OkCupid for women in Los Angeles sorted by match percentage. At the top: a page of women matched at 99 percent. He scrolled down Ten thousand women scrolled by, from all over Los Angeles, and he was still in the 90s. He needed one more step to get noticed.

Women reciprocated by visiting his profiles, some a day. And messages began to roll in. Thought I'd say hi. The math portion of McKinlay's search was done. Only one thing remained. He'd have to leave his cubicle and take his research into the field. He'd have to go on dates.

Sheila was a web designer from the A cluster of young artist types. They met for lunch at a cafe in Echo Park. By the end of his date with Sheila, it was clear to both that the attraction wasn't there. He went on his second date the next day—an attractive blog editor from the B cluster. He'd planned a romantic walk around Echo Park Lake but found it was being dredged. She'd been reading Proust and feeling down about her life.

Date three was also from the B group. He met Alison at a bar in Koreatown. She was a screenwriting student with a tattoo of a Fibonacci spiral on her shoulder. McKinlay got drunk on Korean beer and woke up in his cubicle the next day with a painful hangover. He sent Alison a follow- up message on OkCupid, but she didn't write back. The rejection stung, but he was still getting 20 messages a day.

Dating with his computer-endowed profiles was a completely different game. He could ignore messages consisting of bad one-liners. He responded to the ones that showed a sense of humor or displayed something interesting in their bios. Back when he was the pursuer, he'd swapped three to five messages to get a single date.

Now he'd send just one reply. Want to meet? By date 20, he noticed latent variables emerging. In the younger cluster, the women invariably had two or more tattoos and lived on the east side of Los Angeles.

In the other, a disproportionate number owned midsize dogs that they adored. His earliest dates were carefully planned. But as he worked feverishly through his queue, he resorted to casual afternoon meetups over lunch or coffee, often stacking two dates in a day. No more drinking, for one. End the date when it's over, don't let it trail off. And no concerts or movies. McKinlay's code found that the women clustered into statistically identifiable groups who tended to answer their OkCupid survey questions in similar ways.

One group, which he dubbed the Greens, were online dating newbies; another, the Samanthas, tended to be older and more adventuresome. Here's how each cluster answered four of the most popular questions.

One night. A few months to a year. Several years. The rest of my life. As far as you're concerned, how long will it take before you have sex? I never asked for it. Someone said, "Are you interested in selling? Four months later, we agreed to sell to Match. We felt like it was a chance to give our investors, our employees, and us a great return, and do right by our customers, since Match agreed not to change anything.

For the other 17 people here at OkCupid, nothing has changed. That's part of my job. I have a lot more meetings. I go to Dallas [where Match is based] once a month for two days. I also get to take things off my plate, like finance, HR, and legal, that were less interesting to me.

I like being part of a big company's executive team. It's fun to stretch other parts of my brain, considering questions like, How should we think of acquisitions?

I get to be privy to things that would never come up at a small company. But on the other hand—all I know how to do is start companies. What do I say when people ask me what I do? I mean, a lot of people do. It's great. But it's almost like having a sex-change operation. Part of me is like, What's my identity, if it's not the company I'm building? My OkCupid co-founders are my best friends. We were in each other's weddings. The other night, we went out to dinner. We spent half our time bullshitting around, and then we started discussing our future.