How to use Cohort analysis in Google Analytics

Rather than looking at users as a whole, it breaks them down into related groups for analysis. These groups or cohorts are usually categorised because of some common characteristics or experiences within a defined timespan.

The Cohort Report

Below is an example of what a Cohort Report looks like in Google Analytics:

The Cohort Analysis Report displays the following:

• Cohort type

• Cohort size

• Metric

• Date range

• Table

• Acquisition date cohorts (by metric)

• Segments

Cohort type

This is the dimension that’s the basis of the cohorts. You can only select one dimension at a time. The Cohort Type corresponds to the table column that includes the total number of users in a cohort. For example, if you select Acquisition Date, the cohorts are grouped based on when users started their first sessions. Acquisition Date is the only option available.

Cohort size

This determines the size of each cohort. The Cohort Size corresponds to the date and number of users in each cohort cell in the dimension column. For example, if you select Day, the cells in the dimension column display a single date and the number of users organised into the cohort for that day. If you change your selection to Week, a date range appears, and the number of users organised into the cohort for that week also appears.

Metric

This is the Metric being measured for each cohort. You can only select one metric at a time. The metric corresponds to all columns in the table, except the Cohort Type column, which displays the dimension. For example, if you select Session Retention, each Day displays the percentage of users in each cohort that had a session that specific day.

Date range

The time boundary that determines what data appears in the report. Date Range corresponds to the number of rows in the table. For example, if you select the Last 7 Days, there'll be a total of 8 rows in the table: one for each of the past seven days, and one for the sum of all cohorts.

Table

This represents the metric chosen in the Metric and Cohort Size in the dropdowns above it. Table calculates the average for the chosen period and breaks it down by how they performed through the chosen date range.

Acquisition date cohorts (by metric)

This drop-down allows you to compare up to four cohorts in the chart.

Segments

You can also choose a Segment and analyse the cohort analysis. Do this by adding a segment in the top left of the screen to compare against different segment types.

Benefits of Cohort Analysis

Analyse a group of users and how they're interacting over a period

• Helps identify times when engagement on site drops

• Provides an opportunity to study user or customer churn rate

• Businesses can measure granular customer engagement

• Provides data to calculate customer lifetime value

• Delivers valuable insights into user-retention

To access the Cohort Analysis report, log into your Google Analytics account, and on the left-hand side the Cohort Analysis can be found under Audience Report.

Using the Cohort Analysis report in Google Analytics

The Cohort Analysis report uses two important metrics:

1. Segmentation

2. Per user metrics

1. Segmentation

The Segmentation report provides a view into how a cohort performs each day, week or month after acquisition date. The primary categorisation is the acquisition, which makes it hard to view as it’s an aggregate collection.

The first way to make sense of this is by analysing Segments. This will give you the opportunity to understand exactly which customer type is performing well on the website. For example, mobile traffic, traffic from a specific campaign, logged-in users, unauthenticated users etc. You can also add more than two types of segments to compare best performing customer types.

2. Per user metrics

One interesting feature about the Cohort Analysis Report is its ability to break down metrics per user. Currently the report allows you to breakdown by:

• Goal Completions per user: Total number of goals completed by cohort, divided by total cohort size

• Page Views per user: Total number of views (page or screen views) made by cohort, divided by total cohort size

• Revenue per user: Total revenue generated by cohort divided by total cohort size

• Session Duration per user: Total time spent (hh:mm:ss) by cohort, divided by total cohort size

• Sessions per user: Total number of sessions sorted by cohort, divided by total cohort size

• Transactions per user: Total transactions by cohort, divided by total cohort size

Conclusion

The Cohort Analysis Report gives businesses the power to better understand their user’s behaviour in groups and individuals. This is mainly possible by comparing segments and drilling deeper into understanding per user behaviour and performance for a more advance use compare segments and their individual user performance.

Working with Codehouse

Our Digital Experience Team has helped many customers get started with Sitecore Analytics and Google Analytics. To find out more about how we can help set up Cohort Analysis for your Sitecore website, or if you need help understanding how to get the best out of Google Analytics and other analytics and marketing tools, then get in touch.

Source of images in this article: Google Analytics