Finding love isn’t as straightforward as it used to be and with the rise of dating apps like Tinder and Bumble, the suspense and romance is starting to dwindle.
This is where Crushh comes in… The “anti-dating” app destined to put the spark back into your relationships and uncover hidden romances you never knew you had. We chatted to Founder Es Lee to find out just what this means for the world of dating.
Great to meet you, first things first… what is Crushh?
Crushh is an app that analyses your text messages and uses algorithms to interpret how much people like you. It learns your texting habits and those of the people you text with, and even charts out how your relationships have changed over time.
Interesting! What exactly does it analyse?
The app analyses the SMS text messages and call logs already on your phone (so no additional steps needed to submit conversations). The algorithms try to interpret your conversations like you would, but with better memory and more information. They look at a lot of factors like frequency, length, response times, and features within the texts like sentiment, words, punctuation and emoji usage.
How were you inspired to start the project?
The idea came from a conversation with a friend about dating in New York. He just moved to the city and was confused by mixed signals he was getting from a girl. I looked through the text chain and could tell she liked him. I’ve seen my fair share of confusing texts to start getting a read on “body language”. I figured whatever my thought process was could be converted into algorithms.
Did it take long to get from concept to reality?
We started building in February 2016, so the team has been working on it for about a year and a half. We went through many iterations and tests before launching in May this year.
Over the course of building the app, I’ve learned just how widespread confusion over texts have become. Usually when I tell people what I work on, I’ll find myself interpreting a conversation or explaining the app results minutes later.
What makes Crushh stand out, are there others out there like you?
We haven’t come across anything quite like ours. There are a couple of apps that analyse texting stats and relationships from slightly different angles. We’d actually like to see more apps do what we do because technology should be helping us in all aspects of our lives, especially our relationships. There’s no shortage of apps that connect us with another stranger but too few that help us understand the relationships we already have. In a way, you can think of us as an “anti-dating-app” app.
Who uses Crushh the most?
Teenagers, but we have users across the spectrum. Under 20, it’s more popular with girls and over 20, we have more guys. We suspect it’s because women reach emotional maturity earlier and need a tool like ours less as they get older. Guys will start caring more about their communication as they enter adulthood but need help no matter what age.
Language evolves so quickly, how do you keep up with slang, colloquialisms and differing meanings for emojis?
That’s an advantage of data and algorithms. People can take time to recognise the evolution because they don’t have a perfect view of what’s going on in the world. We, however, factor into the score crowd sourced intelligence on how people text and can help people interpret confusing behaviour. For example, we can see men in their forties use emojis about 1/3 as much as women of the same age. So this difference is factored into the score as soon as the user gives us demographic information. If people start using a certain slang term, we can adjust for that. But this work is still in its early days.
How did you test the app, did you recruit real life daters to analyse their exchanges?
The first algorithms were built on my data. You could imagine this made for some awkward discussions when I asked friends and ex’s whether the algorithms were right. I have friends who now avoid texting me through SMS so they can fly under the app’s radar.
We refined the algorithms on over 200,000 relationships during a year-long beta. Over that period, I got feedback from a lot of self-proclaimed dating experts (a.k.a. singles) who all have their own take on texting. All that collective wisdom has worked its way into the app in one way or another. So has feedback from social scientists, dating experts (practicing ones) and thousands of beta testers.
Users send us thousands of pieces of feedback data every week, so that will help us get better.
Do you think Crushh is better at analysing relationships than another human?
On certain aspects…absolutely. Humans bring their own biases into judgments and our memory is imperfect. In fact, when we ask a friend their opinion about a text exchange, we do so precisely because we know we can’t be objective. Data, on the other hand, is. It can show you someone used to reply to your texts right away now take twice as long.
Also knowledge can be aggregated on a much bigger scale and much faster by a machine than any human mind. A veteran relationship therapist might have seen tens of thousands of relationships over their career. Within the last year, we’ve analysed millions of relationships. And all the knowledge is accessed by our users in a matter of seconds.
That said, we’re still far from replacing human intuition. Humans are still better than machines in most tasks, including understanding relationships. Crushh, as it stands, is simply a tool to help.
What do you think to claims technology, such as dating apps, takes the romance and intrigue out of dating?
We certainly have much more information than ever before. As with any other field, the addition of technology just changes the dynamic. Ultimately people still have choice in how to use technology.
What’s next for Crushh?
We’re just working towards better understanding relationships and communication. There’s still a lot of work to be done on the algorithms and introducing features that allow us to understand our users more. Ultimately we want to share what we learn with the public.
What’s been your proudest moment since launch? And the lowest?
A couple of weeks after launch, we managed to top the dating category charts in the US in every applicable category. It’s pretty gratifying to see our small startup overtaking apps with much bigger teams and large advertising budgets, even if it was just for a couple of days.
The lowest was seeing low ratings from some users and realising we were, and likely always will be, penalised for delivering bad news. Being told someone you really like doesn’t feel the same way about you can be devastating. But we think it’s ultimately better to know and not waste your time.
What tips would you give to other innovators hoping to break the mould?
If you have an idea, just start working on it. Don’t worry about the money, time and whether you can do it. Technology has made it really cheap, fast and easy to start building something. I started with a post for someone to help me build a really simple app for $100. For months, we operated on services that were completely free. Even if Crushh had failed to get off the ground, I would have gained a ton of experience from just acting on an idea.
Download Crushh on Google Play now.