Cohort Analysis in Google Analytics

analytics user cohort

A sample report

This week I was working on creating some cohort reporting for Google Analytics and Mailchimp for a client – cohort reporting gives you great insight into customer loyalty habits and is very powerful and important for start-ups.

In GA I found the easiest way to build automated Cohort reports was via the Analytics reporting API, creating dynamic segments for the month of first visit for a user, and then reporting on subsequent monthly behavior (you could build segments manually in the GA dashboard but it’s not as easy to automate). I’m still working on a Mailchimp version but again will utilise their API and a bit of python!

 

To my surprise today when checking some stats I stumbled upon a new report in Google Analytics does a much better job of this!

Google Analytics Cohort Report

the new Cohort report in GA

 

 

The report is in Beta, and only seems to allow me to look back at 3 months, 12 weeks or 30 days worth of data (depending on if a cohort is a monthly, weekly or daily unit).

 

You can view cohort behaviour such as user retention, pageviews, sessions, transactions of goal completions.

 

All cohorts are assigned by when their first session was, but the tool-tip suggests this is just the first of more to come, so I imagine first transaction or first event would be useful for different businesses.

 

analytics user cohort

A sample report

 

This report sample shows how the cohort data is displayed – unfortunately the account I was given access to the beta through has no traffic so the data is a bit sparse!

 

There is also another report I had not spotted before – ‘active users ‘ – which seems to me to be similar to Facebook’s active user reports – so that you can see of current traffic, how many of your users have been recently active in the last 1, 7, 14 or 30 days.

 

I wonder when it will roll out into my client’s accounts! In the meantime, back to python…

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>