A conversation with Lukas Sliwka, CTO of Grindr at General Assembly, LA.
Last week, General Assembly-LA hosted a conversation with Lukas Sliwka, CTO of dating app Grindr. The topic: managing app growth. To give you a sense of the growth Grindr has experienced, in 2009 it was centered in West Hollywood, California. Now, it has over 2 million unique users per day from 196 countries.
Grindr was originally developed in 2009 by Joel Simkai in his West Hollywood living room and was launched through parties he hosted in Los Angeles and New York. Since geolocation is at the center of the app, having parties is a brilliant launch idea because downloaders could immediately see the core value of the app, providing immediate gratification. Grindr has never taken external investment.
Ably prodded by the unassuming Casey Frith-Smith, VP Video Solutions for Fox Networks, Sliwka delivered a torrent of battlefront advice.
1. Partner where you can for non-competitive advantage components.
To help it meets its objectives, Grindr partners with organizations who can provide technology that is non-core to their technology. So, for example, Grindr has partnered with several organizations to deploy Elastic search. However, they build and manage their own geolocation technology. Since it is so core to the app, innovations they create in geolocation are a competitive advantage.
If you want to partner with Grindr, Sliwka’s is looking to partner with established players only and requires an SLA that provides 24/7 15 minute response. “If you can’t give me that,” he said, “don’t even call.”
2. Security is life or death.
The bigger any app’s distribution gets, the headier the security issues. You have a broader base of people who might want to pull the code apart. But, since Grindr is used in countries where being gay is actually illegal, security literally can be a matter of life and death.
To help shield the data stream from prying eyes, Grindr engages in repetitive encryption all the way down to the device itself. But, Sliwka acknowledges that nothing can be totally secure.
Being an image-heavy app, Grindr faced potential trouble with performance as it grew. So, Sliwka’s team delivers certain “static” components of the app, such as the images, through the Akamai content delivery network (CDN).
To find weak points in the code, Grindr creates logs at different checkpoints in the code that provide a way to report on code performance metrics. For example, logging showed them that an API they were using was taking 1.5 seconds to execute and these same logs measured how changes to the code might help. They brought it down to 500 milliseconds.
4. Be prepared to stop.
I’ve run into this several times myself: the old version of an app/site/product requires this or that fix. If you have identified a new path forward — a path that will take your product where you believe it can go — you have to freeze the old path and dedicate all resources and time to the new path. In the case of Grindr, this required a period without updates to the old version. This was a hard conversation that Sliwka had to have with founder Joel Simkhai and it’s a hard conversation you have to have if you are going to go forward.
5. Hire Polyglots
Sliwka mentioned the idea of polyglots several times and what he means by this is developers who are not solely dedicated to one language or platform. “There are plenty of places for them, just not in my shop.” he said. What does hiring polyglots have to do with app scaling? Grindr is an agile shop that works in one week sprints. That amount of forward velocity requires having people around who can do the work that needs to be done that week. What he prefers is developers who can work as part of a larger team and can lend their expertise on the work that needs to be done.
Here are some upcoming sessions at General Assembly LA:
10/1: Meet and Hire Developers