Mike Ghaffary discusses retention metrics for talent marketplaces, explaining that best-in-class retention plateaus around 70% after 3-12 months, though 30-50% can work if stable. He warns that continuous churn is a leaky bucket problem and emphasizes why keeping employers and demand-side users matters more than supply.
Chapters
So I've been working on a marketplace, right, Mike, you know, which is really, uh, I mean, it's really like a sales talent marketplace, but I'm connecting, uh, companies with fractional sales talent. One of the things that I'm always confused about are the retention metrics. Um, what is a good retention metric, you know, for a talent marketplace?
Yeah, for a talent marketplace or any marketplace, I think the key is that after a certain amount of time, that could be 3 months, it could be 6 months, sometimes it's 12 months, you see some kind of a plateau where one or both sides of the marketplace, you know, supply or demand tends to just stick around and not churn anymore. And so you know that while you have to get all the stuff at the top of the funnel, and some of them are going to churn out, eventually they'll stay. And that can frankly vary quite a bit. The best-in-class plateau you'll ever see is something for like the demand side of the marketplace. Let's say it's involving, you know, and for you it might be supply on a labor marketplace, but one side is more churning, right? And so for talent marketplaces, obviously that's usually more the— that's probably more the supply side. In theory, the people who need the jobs are kind of consistently going on. You know, that could change in a recession. And so for your more churning side, 70% plateau retention of the original cohorts is like best in class. But frankly, even 50, you know, even 30 or 40% could be a great plateau if you can show that it really stays level after that. Now, on a talent marketplace, ultimately, your, your supply could just turn over quite a bit. You could say, look, you know, the— we have one-and-done type people come in, Maybe they reactivate a couple years later. I think what you have to show is that your demand side, the employers, that they really stick with you. So I think if your plateau retention on the demand side, the employers, is too low, or if it doesn't plateau, let's say you just consistently churn anywhere from 3 to plus percent, 3 to 10% or more, But even at 3% or 4% a month of the employer side, and that keeps going down month after month after month, and there's no end in sight. You're still 12 months out churning through. The problem is investors are going to say, this is a leaky bucket. We actually have no advantage. We eventually, fast forward a year or 2, have lost all of our supply and have to refill it all. And the demand was the more churning side. Supply was supposed to be the more stable side. Um, so that's— so you can apply this logic to any of your marketplace, you know, even if it's not labor marketplace, talent marketplace, even if it's reversed. If your more churning side can't plateau, that's a problem. Uh, or sorry, if your less churning side can't plateau, that's a problem. And ideally you can show like what Uber and Airbnb could show was that they had some stickiness on both sides.
