Cohort analysis
The Cohort analysis widget provides a high-level view of cohorted revenue activity for your campaigns. This lets you quickly check if your cohorts are performing as expected. With these insights, you can optimize your campaign strategy.
Use data from this widget to:
- Measure user retention
- Estimate an app user’s lifetime value (LTV)
- Compare the in-app performance of new installs against your reattributed users
- Analyze user behavior before and after app modifications
Your widget displays data from the following sources, by default:
| Data source | Definition |
|---|---|
| Attribution source - First | User's original attribution source |
| Attribution status - All | Installs and reattributions |
| Attribution type - All | Clicks and impressions |
| Ad spend source - Mixed | Attribution and Network sources |
Set up the Cohort analysis widget
Before you set up this widget, ensure that you have set the appropriate filters. For more information, see Set up your view.
The Cohort analysis widget is a table that displays the performance of the selected cohorted metric by the dimension.
By default, you can see the analysis for 0D All Revenue Per User by Day of install. You can group your selected metric by dimension. Select the Dimension dropdown to see the list of available dimensions to choose from.
Use the Metric dropdown to choose the metric for which you want to analyze your cohorts. Certain metrics need to be further defined, to determine how Adjust calculates them for reporting. For example, for revenue metrics you need to set a revenue type. Choose additional options for the metric, if applicable, and choose Select Metric.
Select Days, Weeks, or Months to choose how granularly you want to break down the data. Depending on what you choose, your cohorted data can display for different available time periods.
Examples:
| Selected time period | Nth day breakdown | Cohort periods in table |
|---|---|---|
| May 12 to May 22 (11 days) | May 12 - D0 May 13 - D1 May 14 - D2 May 15 - D3 May 16 - D4 May 17 - D5 May 18 - D6 May 19 - D7 May 20 - D8 May 21 - D9 May 22 - D10 | D0, D1, D2, D3, D4, D5, D6, D7, D10 |
| May 12 to May 23 (12 days) | May 12 - D0 May 13 - D1 May 14 - D2 May 15 - D3 May 16 - D4 May 17 - D5 May 18 - D6 May 19 - D7 May 20 - D8 May 21 - D9 May 22 - D10 May 23 - D11 | D0, D1, D2, D3, D4, D5, D6, D7, D10, D14 |
| May 12 to August 3 (12 weeks) | May 12 to May 18 - W0 May 19 to May 25 - W1 May 26 to June 1 - W2 June 2 to June 8 - W3 June 9 to June 15 - W4 June 16 to June 22 - W5 June 23 to June 29 - W6 June 30 to July 6 - W7 July 7 to July 13 - W8 July 14 to July 20 - W9 July 21 to July 27 - W10 July 28 to August 3 - W11 | W0, W1, W2, W3, W4, W5, W6, W7, W8, W9, W10, W15 |
Select 
(Chart view) to view the analysis in the form of a line chart. Select
(Table view) to switch back to the table format.
Select 
(Open as report) to view the data in the form of a new report.
Use the Cohort analysis widget
Hover over any data line in a chart to see more information about the data.
Hover over any title in the chart legend to highlight the data line in the chart that corresponds to that title.
Select the title in the chart legend to hide the data line for that title in the chart. Select the title again to reveal the data line.
