Create a revenue team with its own iOS and Android engineers

Jeff Morris… And I started to put my revenue hat on and he was like, "What are you talking about?" And I was like, holy crap, you hit 50 million without any experience. I'm sure if you gave myself or anyone who loves revenue just the keys to the application for a couple sprints, we could probably double your revenue or do something that would really shock the team. And so it kind of reminded me that the sophistication we got to at Tinder took us a long time.

Jeff MorrisWhen we really built out the revenue team, it was 2016, about a year into my time at Tinder. And it was just myself and call it five or six engineers. You had three iOS engineers and three Android engineers. And then leaving the company, I think I had 40 plus engineers who were working on all parts of the revenue roadmap. And I think that's kind of where a lot of these AI companies are in their product journey. So it's almost, as we say that, a lot of them are trying to monetize really early and then they kind of do this set it and forget it thing where they're not really refining that part of their product. So I think it can lead to, again, even more false positives or metrics that might not prove to be durable over time.

David BarnardWell, without revealing too much about that startup, can you share what your first two experiments would be? I mean, did you go through the product yourself and say, hey, here are the top two I would run?"

David BarnardYeah. And in our notes that we're working on ahead of the podcast, you wrote something about that I thought was really great. It was, "We spent too much time building Excel features instead of testing ideas." And I assume that's what you meant, you overthink things and run the models and price everything out and make assumptions and overthink instead of just getting out there and testing it. Is that what you were talking about?

Jeff MorrisSo we were a private company and then we became a public company and Tinder itself was 90% of the revenue of that public company. And so it was actually a ton of pressure on us to perform every single quarter. So you basically had 12 weeks where you had to deliver some revenue growth. And if you didn't do that, literally the stock price would go down and everyone was mad. I think the Excel version of this was we just had more and more finance team members who wanted to help us forecast the impact of what we were doing so then they could go to the next earnings call and give what would be a reliable forecast. But that created, I think, a bit of Excel culture where the product team was forced to spend a lot of time speculating on things that were hard to predict.

Jeff MorrisOr we'd have things that would happen in the quarter that were small parts of our roadmaps going into the quarter that would work in a really big way that would make up for any kind of inconsistent forecasting. So that was the biggest thing was we kind of... I think as your company gets bigger, the org gets bigger, you become a public company, the stakes get higher and so you have to do things with more precision and in some ways that can slow …