A Tinder Tool: Ghostbusters!
Who you gonna call?
Today I want to present to you all my fastest lightning project, a Design Sprint that we carried out in less than two hours. In addition, it was the result of an open workshop in which I teamed up with professionals in the UX field that I didn’t know at all.
It was the shortest Sprint I’ve ever done and, nevertheless, it’s probably the one I’ve enjoyed the most. Why? Because there was no time to hesitate and, in some way, that hyped up my creativity to the top.

The lightning Design Sprint
To face this challenge, it was important that the team followed a very clear plan. First, we had to establish the challenge. We started out with one rule: we had to add a feature to an existing app. We thought the Tinder dating app was a fun one, especially since we had a couple of boomers on the team who didn’t really know how it worked.
To spice things up a bit, we decided that our challenge was to add a feature to the Tinder App using Artificial Intelligence.
Sprint Questions
Every UX designer knows that the first thing to do in a Sprint is brainstorming questions and dividing them into themes; this is how our Sprint Questions are born. It had to be a very quick brainstorm, so we got to work, merged the repeated questions and classified the most interesting ones:

User persona
Then, we designed a user persona (very basic, due to time constraints) trying to gather the needs of an average Tinder user and the steps she usually follows when using the app, as well as her main goal: connecting with someone.

“How Might We…?”
The next step was to turn all the questions into “How might we…?”, or more specific and accessible questions to address the overall challenge. Then, we voted on the ones we found most interesting. While “How could AI detect narcissists and psychopaths on Tinder?” was a huge success (for whatever reason), the winner was “How could AI report the percentage of ghostings a user has ever done?”.

Lightning demos
Come on, quickly! The clock is ticking!
In a Design Sprint of this sort there’s no time to lose, but it is also true that there’s no UX project without its benchmarking, even if it’s a little one. And this was the moment to do ours:

We found a lot of information in blocks of text, so we introduced small notes summarizing what we had found on post-its, so that our discoveries would be easier for our colleagues to read. Teamwork!
Crazy 8
Crazy 8 time! We grabbed our post-its and started landing ideas, eight in total, one idea per minute. Tick, tock!

Time! We couldn’t write any more (otherwise, we would’ve keep proposing things, and we had a lightning deadline).
And now, what’s our favorite idea?
Prototype
There were two finalists: “a feature that makes AI kindly leave you if it notices a user has ghosted you” and “a feature that detects the risk of a user ghosting people and warns about it on their profile.” The team chose the second one unanimously.
Thus, we got to our prototype: Ghostbusters!
How does it work? AI analyses users’ behaviour on Tinder and detects which profiles stop replying after having a good number of interactions with other users. The result? A notice on their profile, which they cannot see nor delete, alerting potential matches. More or less, like this:

And here is where our Design Sprint ends, because we ran out of time and because we have our new feature for Tinder all sorted out. And because, in the quick testing we did at the end, our friends told us that it would be a very useful and funny warning to tease the ghosts. So the team proved this feature a resounding success.
Who you gonna call? Ghostbusters!