Like the rest of the world I have been reading the Lean Startup, the new book from serial entrepreneur Eric Ries. There is a lot of good advice in the book, which focuses on developing products which solve real problems and doing it in the most efficient way.

But one idea really stood out for me: cohort analysis.

This is one if the most important tools of startup analytics. Although it sounds complex it’s based on a simple premise. Instead of looking at cumulative totals or gross numbers such as total revenue or total number of customers, one looks at the performance of each group of customers that comes into contact with the product independently.

This is a powerful idea. Often there is a lag in the behaviour you would like to see (such as signing up for a free trial) and this is very hard to detect in a mass of other numbers, which may all be going up. By concentrating on one group at a time, Ries argues, you can really tell if the key behaviours you need to see are really happening – and if you are getting better or worse over time.

If you are interested the key section starts on page 123 of the book, or you can find out more from his site. More on cohort analysis here.