Track what Matters with Cohort Analysis

Ever since reading Eric Reis’ The Lean Startup, I have realized the importance of using cohort analysis to track the actual growth of Purlem.  For the last few years in Purlem, I have been nearly 90% in development mode.  My head has more-or-less been in the sand as far as growth, and how it happens.  I’m happy to say, that is all about to change.  Here is how I’m using cohort analysis to effectively measure Purlem’s growth.

First of all, a brief explanation of what cohort analysis actually is:
A cohort is a group of people who share a common characteristic or experience within a defined period

Eric Reis calls cohort analysis one of the most important tools of startup analytics.  “Instead of looking at cumulative totals or gross numbers such as total revenue and total number of customers, on looks at the performance of each group of customers that comes into contact with the product independently,” Reis says.

With Purlem, I’m using cohort analysis by grouping users by the month the signed up.  For each month I track the following:

  • New users
  • Users that created a campaign
  • Users that uploaded contacts
  • Users that received results
  • Users that paid

So what is all of this telling me?  For starters I can see how the improvements to Purlem’s platform affects usability.  For example, notice the numbers of users that created a campaign (red line) vs the number of new users (blue line).  Back in 2010, say about 20% of users actually created a campaign.  Today, approximately 80% of new users create a campaign.  I attribute a majority of this growth to the instantPURL feature we released last summer.

I can also see that there is a sticking point to users actually adding contacts – something I will need to give more attention to moving forward.  Finally, and quite worrisome, is that paying users far out way users that actually added contacts and received results.  So people are paying for Purlem, but not actually using it.  I’m not quite sure how to think about this yet, but again, something I need to give more attention to.

Before using cohort analysis, I was tracking the cumulative number of paying users.  Eric Reis calls this vanity metrics as they give the “rosiest possible picture” of a startup’s progress, but does not track how people are actually interacting with the application.

At the end of the day, using cohort analysis helps you to track the numbers that matter to the progress of your company.