Understanding how your ad spend performs over time is critical for optimizing your marketing strategy. With Cometly, you can easily build a Cohort Analysis Dashboard to track how specific groups of users—acquired during a defined time period—convert or generate revenue over time.
What Is Cohort Analysis?
A cohort is a group of users who share a common characteristic, such as signing up during the same month. Cohort analysis allows you to track the performance of this group over time—for example, how many of them convert to paying customers weeks or months after initial acquisition.
Step-by-Step: Building a Cohort Dashboard
1. Create a New Dashboard
In your main Cometly dashboard, click the down arrow beside your dashboard tabs.
Select “Create New Dashboard” and give it a name like “Cohort Analysis.”
You'll start from a blank dashboard.
2. Add Your First Report: Ad Spend
Click “Add Report.”
In the report builder, choose “Spend” as your primary metric.
Set the date range for your cohort—for example, April 1st to April 30th.
Choose a line graph, table, or KPI summary—whichever you prefer for visualization.
Click “Add to Dashboard.”
This gives you the total ad spend for acquiring users in the selected time window.
3. Add Your Key Conversion Metric
In this example, let’s say your primary conversion is Free Account Signups.
Add a new report to the dashboard.
Set the date range to "After April 1st" to include all people who upgraded during or after the month of April
Choose the early funnel event from this cohort that you want to see longer term performance for, like in the video above we use “Free Account Signup” as the example event (ensure the event you choose is already set up in your Events Manager).
Add a filter to ensure you're only including signups that happened within the cohort window.
To filter:
Search for the event field, in our example we would search “Free Account Signup Date.”
Set the condition to “is between April 1st and April 30th.”
Click “Done.”
This ensures you’re only analyzing users who signed up during the April cohort.
4. Track Post-Signup Conversions
Next, you want to measure what these users did after signing up—did they upgrade?
Add another metric: Upgraded to Paid Subscription (once again, this would need to be an Event that you have already configured in the Events Manager) and Gross Revenue (or the correlating revenue metric for the upgrade event that you are measuring)
💡 Many users sign up during a cohort window but upgrade weeks or months later. Cohort analysis helps capture this delayed ROI.
5. Attribute Revenue Correctly
Cometly gives you control over how revenue is attributed.
Set your Attribution Window to Lifetime Value (LTV) to capture all revenue generated over time by this cohort.
You can also explore several advanced attribution models to understand what sources drove signups.
Switching between attribution models can provide strategic insights into which touchpoints matter most during the journey.
6. Repeat for Future Cohorts
To build a cohort for May, simply:
Clone your reports with updated names, and date ranges:
Change the spend report name to "May" and change the date range to May 1st to May 31st
Change the name of the cohort report to "May" and change the date range in the cohort report to After May 1st
Also change the signup date range in the filter section to between May 1st to May 31st
Click "Add to dashboard"
Keeping each cohort in a separate dashboard allows for clean comparisons month over month.
Summary: Why Cohort Analysis Matters
Cohort analysis helps you:
Understand the true ROI of your ad spend over time.
Spot delayed conversions that happen weeks or months after acquisition.
Compare how different monthly cohorts behave and convert.
Refine your ad strategy based on long-term value, not just short-term wins.
By building a cohort dashboard in Cometly, you’ll gain clearer insights into which campaigns and audiences deliver the best return—even months after the spend happens.
Need help? Reach out to our Customer Success team or watch the full training video at the top of the article.