​
Overview

User retention is critical for the success of digital products, it’s linked to the product’s ability to consistently deliver value to the users and is a key measure of business health. It is also essential in the process of achieving a product-market fit and retaining constant growth. Definition: Retention is designed to assess user engagement over a specified period of time. Our new Retention Analytics processes data into an easy-to-understand format, where you can see how much your users interact with your product.

​
Use cases

For better illustration, we’ll consider a customer who manages a messaging app.
  • On average, how many users are still active after two weeks from signing up?
  • What percent of all users are still sending messages after seven days?
  • How has my 7-day messaging retention changed over time?
  • What percent of users sent messages in 2, 3, or 4 distinct hours of the day?

​
Creating a Retention Report

To create a Retention Report, click on Analytics on the dashboard, then click on Retention. You will see a view split into two sections:
  1. Left Section (Query Builder): Here you can select the data (events and properties) and filters you wish to apply to the report, it consists of two tabs: Events and Filters.
  2. Right Section (Results and Visualization): You can review query outcomes and choose data presentation options.
Note: You can view the list of Retention Reports you have created on the Saved Reports page.You can also create a new Retention Report on the Saved Reports page by clicking on the button Create Report.

​
Events

Step 1: Select Users or Companies Here you can choose whether you want to analyze the data on a user or company level. Step 2: Select ‘Who Performed’ Event
  • You can define the initial event from which retention begins. You can select events from three categories: Features & Events, Pages, and Content Engagement.
  • You can also decide whether you want your Retention report to consider either:
    • For the First Time: Only the initial occurrence of the event.
    • Recurringly: Include subsequent occurrences.
Step 3: Select ‘Then Returned to Perform’ Event Similar to the previous step, here you can select the event you wish to consider for retention. Step 4: On Any of the Next This is the Retention time frame. It is possible to set the period (days, weeks, or months) over which retention is measured. This will tally users returning in the succeeding interval. Optional Step: Inline filters Inline filters allow you to filter each event separately (as opposed to the Filters tab, which will filter all your selected events at once). The inline filters option appears once you hover over each event. Events can be filtered based on:
  • Events Properties (also called event attributes)
  • User Properties
  • Company Properties

​
Filters

You can filter all your selected events at once by clicking on the Filters tab. Here you can filter them based on User Properties or Company properties to drill down deeper into the analytics.

​
Results and Visualization

Once you click on “Run Query” you will be able to see your results on the right section of your screen. First, you can select the Date Range in which you want to measure your Retention Data. There are different two types of data visualization options, it is possible to view this retention data in a Cohort Table or as a Trend.

​
Cohort Table

Cohort column is shown in green and the Average Retention row shown in red A) Cohort This indicates the time period (day, week, or month) when the user performed the event and the count below indicates the number of users who performed the event within that time period. The remaining columns to the right are the ones displaying retention. B) Average Retention This data appears on the first row and is the average across all of the cohorts’ boxes for each day, week, and month in the table. C) Example If you measure weeks, the ‘< Week 1’ column represents users or companies that have completed both events in this period before ‘Week 1’. The following columns will show you how many of the users who performed the selected activity on the week marked under ‘Cohort’, came back and visited again performing the selected return activity on Week X. Notes:
  • User cohort buckets count unique users and companies (not event totals) who met the event criteria in a specific time window, with each customer counted once per bucket but can be included in multiple buckets.
  • For example, if you are bucketing based on your “Item Purchased” event and creating weekly buckets, a customer who purchased at least one item each week will be in every bucket, not just the bucket of their first purchase.
  • Cohorts are added and removed on a First In First Out basis. They are added at the bottom of the table daily, and the first row (oldest) is removed from the top.

​
Trend

The Trend view is a line graph of the average retention row in the table view. For instance, if the table shows the last seven days, the graph will also show over seven days with each point corresponding to the percentage for that day. This will function the same with other time periods such as weeks and months. You can choose between Linear or Cumulative.

​
FAQs