Cohorts, for those that aren’t familiar with them, are groups of users who sign up during a given time period (say, everyone who signed up between January 1 and 7, 2011). And analyzing them can offer some valuable insight into how people are using (or have stopped using) your service. Kissmetrics cofounder Neil Patel explains that the vanity metrics we often see being discussed in the press, like total user count, can be misleading. For example, your site might have growing overall usage each month, but that growth may be fueled entirely by a rise in new signups — who could mask the fact that your older users are fading away.
That’s where cohort reports come in handy: they’ll let you break your overall user base into different groups, then see how each of those groups use the service during, say, the following six months. This can be helpful both to gauge the lifetime value of users, and to see how much of an impact new features are having on each group of users. Kissmetrics will also let you further break your users out into cohorts based on how they were referred to your site.
Patel acknowledges that there are other services that have set out to offer Cohort Reports as well, including Mixpanel and RJMetrics, but says that their products, at least as far as cohorts are concerned, give users less flexibility (he also believes Kissmetrics is easier to use). Patel says one key feature offered by Kissmetrics that the others don’t have is the ability to ‘zoom in’ on each cohort down to the individual user level.
Kissmetrics is available for free for a 30 day free trial, with premium plans available after that beginning at $29/month (prices increase depending on the size of your size and how many datapoints you’re tracking).
If you’re interested in learning more about cohorts, Evernote CEO Phil Libin has previously spoken in detail about how his company uses this kind of data.