The Industry Got It Wrong

How to build the most advanced algorithm ever!

I would really love to tell you we are building the most advance algorithm to match people with the best possible dates ever but honestly, we are actively doing the opposite. Such a technology is only as good as the amount and quality of information we feed it and let’s be honest, we hate filling forms.Then again, society doesn’t actually work that way. Life doesn’t filter people we meet on a daily basis based on our behavioral traits or preferences. This is not to say nature doesn’t provide some sort of filter at all, it does. This filter is mostly based on our everyday life experiences such as friends we encounter and things that affect us in our immediate environment. For instance, it is safe to say that majority of our friends share some common traits with us, i.e. age, taste and preference, location, education, work and so on.

Instead of reinventing the wheel by building a match making algorithm like the top dating apps (eHarmony, Zoosk, Match.com, OKCupid), we are simply adopting the commonest and MOST successful way we meet new people; through friends. Show me your friends and i’ll tell you who you are. But wait, we are not using your Facebook social graph either. Well, the closest representation of your physical social network, resides in your phonebook. The phonebook is the oldest and most powerful social network even before we gave birth to the internet and the social media bubble. If you have ever kept a physical phonebook of your friends’ telephone numbers, you’ll understand this much better because losing that phonebook back then was equivalent to losing your Twitter, Facebook and Instagram logins all at once.

Now your phonebook is on your mobile phone and therefore your friends can be with you on any app, ask Facebook.

We want users to simply sign up and get authenticated using their phone numbers instead creating a user name and password or Login through Facebook. Once the user is verified, we can use their address book to find their friends who are on the app and vice versa. This process makes most sense to us because as much as we love for users to meet new people, we want them to meet them through friends. So connecting users initially with their friends is as important as meeting the third party.

We are not claiming that current online dating is not helping users meet new people or not profitable in that regard, after all it is a billion dollar industry. So before we go any further, let’s take a look at some of the industry’s TOP players based on their technology and why we chose to go the other way in each case.

Algorithms and matchmaking technology (eHarmony, Zoosk, Match.com, OKCupid e.t.c)

Using technology to help users find love, long term relationship and marriage is the common objective of dating sites and apps such as Zoosk, eHarmony, OKcupid, chemistry.com and match.com. Each dating site’s algorithms may differ slightly but the underlying technology is the same. You answer a series of questions and the program matches you with carefully curated potentials based on your responses. The program defines users taste and preference by what they respond to. It is a continuous learning process, that records the users responses to major characteristic differences and follows up later with other variables to double check whether it was one characteristic the user was preferring over the other.

This category are the most profitable services in the industry due to their long establishment and paid subscription models. They claim to have largest number of singles, with millions of users investing considerable time and money on membership each year to have access to potential dates.

This sounds awesome, so why not jump on that bandwagon and enjoy the wave? Well, technology can’t be the sole solution to the search for love and relationship but if we are to involve software, we might as well start from a credible source of data of users, FRIENDS. A system that can warrant a safe off-line meeting. Saying we have a certain number of singles on our service is just not enough. Who are these singles, how did we acquire their data, where do they live, who knows them, was in school with them or works with them? Most importantly, how do we verify their identities with ease?

It is easy to manipulate the total number of users like most dating sites do, and brag about millions of singles in a few months. There are several ways to cultivate fake profiles including automated profile generation sites such as http://randomuser.me/. With a combination of actual user sign ups and a couple of fake profiles, any dating site can easily overwhelm users with thousands of potential dates.

Fake Profile Example
Randomly Generated Profile from http://randomuser.me/
Location Based Dating Apps (Tinder, Sway, Single Around Me, Hot or Not e.t.c.)

Location based dating apps are fun and addictive ways of meeting new people. Most of the location based dating apps have adopted the card swiping technique popularized by Tinder. Cards provide bite-size data preventing information overload and are great for making decisions about now. Other apps such as Jelly, Swell, Ness and recently Paper from Facebook is also experimenting a news feed with a card interface. For a guy, I couldn't have asked for something better when I first saw the double opt feature on date cards in dating apps. With very little concern about personality, I simply keep swiping right on as many good looking cards as possible till I get lucky with some random girls around me. Well, only proximity is somewhat achieved here without addressing the fundamental problems of meeting new people; safety and verification.

Women especially want verification, authentic, ease and a safe offline meeting. Location based apps do not guarantee these things.

So, how do we achieve proximity, verification, ease of use and safe offline meeting all in one place? Well, we safely lean on friends one more time. The people we talk to often mostly likely live around us. These are are friends we hang out with in the neighborhood, on campus, at parties, at work or at the games. Using phone contacts, we can verify users and allow them to get to know friends of friends easily. This cuts location barriers while ensuring a safe and creep-free environment for all.

Facebook Based Dating Apps(Hinge, Down, TrintMe e.t.c.)

In addition to the usual question about how you're going to approach dating differently, you have to answer the even more important question of how to overcome the huge chicken and egg problem every dating site faces. A site like Reddit is interesting when there are only 20 users. But no one wants to use a dating site with only 20 users — which of course becomes a self-perpetuating problem. -Paul Graham

We spent the better parts 2011 and 2012 building social media strategies and Facebook apps for several corporations. Facebook became our home; we built from Gaming to Customer Service apps on the platform. With our first ever Facebook project, the Tigo Ghana Facebook page grew from Zero to over 30,000 Likes using our native Photo Contest App in less than a month. Our results as pioneer social media agency in Ghana was so compelling that we got hired by our clients competitor a year later. We saw organic fan growth in the hundred thousands in subsequent Facebook projects.

These achievements were largely based on Facebook’s expansive social graph.

For dating apps however, Facebook connect remains a means to an end not the curtain call.

It may be a great source of user acquisition through ads and fan engagement on the dating apps page but would not entirely solve the dating industry’s problem of user base.

Fake Profile Example
Hinge Facebook Login

However, dating apps such as Down and Hinge that primarily adopt Facebook’s social graph may give users the wrong impression. A user’s massive Facebook friends network on such dating app doesn’t mean all of those friends are using the app as well. Facebook logins for such apps can not also be used to create virility since users login through Facebook under the premise that the app would not post on their wall or spam their friends.

Phone number sign ups don't need to reassure it won’t post on a users behalf. There can also be loads of fake Facebook accounts or Facebook friends that we are not actually friends with, however, phone contacts would allow us to only find real friends of the user who are also on the app and create a more meaningful connection.

Conclusion

Algorithms, locations, email signups and Facebook are the commonest components of all dating sites. However, an important change is not necessarily about following industry standards. It’s about paying attention to details that really matter and dropping off unnecessary load. The sum total of what we want to achieve consists of what it isn't as much as what it is; Simple, Safe and Fun way of meeting new people through friends.

We are currently testing Beep on iOS and Android for a release in February 2015, you can check out our website or catch up with us @OfficialBeep on Twitter. Better yet just click the recommend button and be part of our story!

Share this article on:

Blog

Our thoughts are fucking awesome so read the damn shit!