Patook wants to be Tinder for platonic friendships

Technology isn’t necessarily the best solution for forging new real life connections, whatever the collective marketing spiel of the web’s social giants likes to claim.

An especially tricky type of human connection for an app to ‘fix’ is making new friends, as friendships are often forged in the fire of random real life experience. And while there’s no shortage of out-and-out dating apps, offering ample opportunities for hook ups with individual/s from your sex of choice, what if you just want to make new friends to hang with platonically?

Patook is an app that’s been around for a little while (the beta launched in mid 2016) which is attempting to crack this tricky feat of remotely sparking friendships. It bills itself as the “strictly platonic friend making app”.

Its bootstrapping team officially launched it out of beta this month, and it has around 70,000 users at this stage, who founder Tony Daher says are sending around 15k messages per day.

Somewhat ironically, Patook recycles the Tinder ‘swipe to like’ mechanism as one of the options for screening potential friend matches, although you can also switch to a list view — arguably a better fit for a friend-finding scenario where making snap judgements on whether someone is potential friendship material just from a short online profile seems even less sane than doing so in a dating context.

Before you get to that bit, the app has a fairly customizable filter interface which allows you to spec out some basics for finding potential buddies — offering the ability to select gender; whether you want to friend couples; specify an age range; and set how far from your location to search.

There’s a profile with various likes/dislikes fields that can be filled in, and users can also take a questionnaire to further flesh out what the app knows about them so it can better tailor matches.

If you take the (optional) questionnaire the app will use your responses to hone its core matching rules, so it might, for example, promote matches with people who have kids, or who hold liberal views, or who don’t smoke, and so on.

Patook’s flagship tech trick is one that aims to underscore its friends-only mission: it’s using Natural Language Processing to help power a “flirt detection feature” aimed at preventing users from straying over the platonic line and trying to hit on others by sending suggestive messages.

Now you might think just calling your app a “strictly platonic friend making app” is enough to prevent people getting the wrong idea. But Daher isn’t so sure.

“There have been some attempts in the past to create friend-making apps but they’ve turned into dating apps after short periods,” he writes in an online explainer. “The premise of Patook is based on eradicating that problem.”

“Anything that is even a hint more than strictly platonic is immediately banned. No romantic advances, no flirting/hitting on, no innuendos, no “friends first then we’ll see” behaviors… Over 5% of users who tried to join were banned before their first message was even delivered.”

At this point it seems worth noting that dating app behemoth Tinder does itself have — what’s billed as — a “friend-finding” feature, called Tinder Social.

However, on closer inspection, the actual use-case for Tinder Social is probably charitably described as ‘group dating’. (Vice’s conclusion after using it for a month? “It’s definitely for gangbangs”. So, er, still ‘dating’ then.)

So, well, you can see the problem with using services that are overarchingly known for dating and/or hook ups for anyone who genuinely just wants to find new friends.

And, if you’re in a relationship, just having the Tinder app on your phone can be a conversation starter of the wrong kind with your S.O.

The long and short of all this is, anyone who’s interested in meeting new people for actual coffee and genuine chit-chat probably doesn’t want to do so within the wrapper of a dating app.

Daher says Patook built its flirt-detecting, messaging parsing algorithm by training it on sentences considered flirty (some culled from Reddit) and others considered normal.

“This was done by crawling some image submissions from various sites, and running them through Optical Character Recognition,” he says. “If you go to say reddit.com/r/niceguys or reddit.com/r/creepypms you’ll see a bunch of ‘bad’ examples posted as images. We picked some of them up and translated them to text.”

The app also performs ongoing behavior analysis to further weed out any creeps — such as looking at the kind of people a user is messaging; whether they are sending pictures only; whether they’re repeat sending the same message etc etc.

“All the above is put into a Bayes probability classifier that keeps a working probability of a user being there to flirt,” he adds.

TechCrunch downloaded the app to test it out, and after about a week of (fairly light) usage I can safely say I have received no unwanted advances, neither in photo nor text form. Indeed, I’ve not received any likes or messages at all (I’m not counting the message from the app admin who noticed I was a journalist and wanted to see if he could encourage me to write about the app); and just four visits in total to my profile.

And so we come to what looks to be the main problem for Patook: convincing its — likely — typically shy target demographic to really put itself out there and actively solicit attention from strangers within such a blatant “friend-making” context.

Asked what barriers he sees to people using the app to make new friends (over and above the creep factor it’s actively working to mitigate), Daher agrees there are a “bunch of problems” — including people not yet being comfortable with the concept of a friend-making app and the fact that, therefore, “very few people talk about it”.

“This is reminiscent of online dating a decade ago where it was nearly taboo to be on such an app. I believe the tide is turning and people will change there,” he suggests.

In terms of specific steps Patook is taking to try to warm things up and get users chatting and virtually bonding, he says they’re dabbling in gamification for starting conversations.

The app also rewards users who maintain long conversations with people they’re interested in — to try to get around the problem of people “dropping out” of conversations.

Although, in my view, Patook’s points-structure also works as a disincentive, by requiring users earn and spend usage points to do things like boost their profile or to view all visitors who’ve liked them. Point being: If you’re starting from a position of there not being enough users in the first place, requiring your (very small) pool of users to earn and spend points to do things like boost their visibility seems a bit pointlessly heavy-handed (even if it’s intended to encourage greater usage overall).

Daher agrees that lack of users generally is a big barrier for Patook. In practice this means there’s unlikely to be enough relevant matches in the same city for most users to feel really compelled to start conversations. In my case I liked a few other profiles but never felt an urge to proactively try and break any ice.

“Except for the major hubs most cities only have a few hundred users who know of the app right now so that makes it less likely for someone to find a genuine connection other than ‘they just live in my city’,” he concedes.

Discussing the users it does have, Daher says the “main” demographic so far are women in their twenties or thirties. “We also have a lot of ‘we just moved to this city’ people,” he adds. “And the ‘got married and need to find someone else who has children to hang out with’.”

In terms of successful matching, he says some of the people for whom Patook works best tend to be those with “rather unique interests” — given that the app matches rarer interests first. “So people are happy to find someone who’s also into that,” as he puts it.

Overall, while I like the idea, and appreciate Patook’s attention to detail in building anti-creep tech, as it stands it’s hard not to conclude this friend-finding platform is most likely to leave you feeling a bit, well, friendless. It doesn’t take long swiping through the slender selection of profiles in your own city before you’ll be reading about the interests of strangers living in other countries, and wondering what the point of the app is.

If the future of work means a lot more Internet-connected remote workers who aren’t regularly sharing a physical office space and thus not benefiting from easy access to a workplace social circle, there arguably is going to be a need to find alternative ways for people to forge new friendships. I’m just not sure another app is the best answer to that, however honorable its intentions are.