Creating Cohorts Based on Selected Time Period in Dashboard
Is it possible to make cohorts relative to a time period selected in a dashboard?
Is it possible to make cohorts relative to a time period selected in a dashboard?
Hi Jalil K., Please stand by as I search the Mixpanel knowledge base for an AI-powered answer. I normally take about 15 seconds or so. In case AI cannot answer your question, feel free to reach out to Mixpanel support at mixpanel.com/get-support. If you have follow-up questions for this thread, please tag me again with your question: @kapa.ai <follow-up question> đź‘‹
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
Mixpanel cohorts are computed dynamically at the time you run a query, meaning the cohort always reflects users who meet the cohort criteria as of the current moment—not relative to a time period selected in a dashboard. The set of users in a cohort is not a rolling window of users who have ever been in the cohort, but rather those who are in the cohort "right now" when the query is executed. This means that if you change the date range in a dashboard, the cohort itself does not automatically adjust to reflect users who met the criteria during that selected period; it always uses the current cohort membership at query time How Cohorts are Computed.However, you can define cohort criteria that reference relative time windows (e.g., "users who signed up in the last 7 days"), but the evaluation is always based on the time the report is run, not the dashboard's date range Creating Cohorts.In summary, Mixpanel does not support making cohort membership dynamically relative to the dashboard's selected time period; cohort membership is always computed as of the current time.
kapa.ai is still learning and improving, please let me know how I did by reacting below.
ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.
Yes — it’s absolutely possible, but it depends a bit on the BI/analytics tool you’re using. In general:
Cohorts are usually defined by the first time a user (or entity) does something (e.g., sign-up date, first purchase date).
To make them relative to a time period you select on the dashboard, you’ll need to filter or re-bucket the cohorts based on that selection.
How this works in practice
Define the cohort event → e.g., “user signup date.”
Apply the dashboard date filter → instead of calculating cohorts for all time, the query respects the dashboard’s selected range (e.g., Jan–Mar signups only).
Group by cohort → each user is bucketed into a cohort based on the event date (within the selected period).
Measure retention/behavior → plot how those cohorts behave over subsequent days/weeks/months, relative to their own start date.
Example
If your dashboard filter is March 2025, the cohort logic will only include users who first did the event in March 2025.
The retention curves (or whatever metric) will then be shown just for those March cohorts, not for February or earlier.
Tool-specific notes
Looker / Power BI / Tableau → Yes, you can parameterize the cohort start date relative to a dashboard date filter.
Mixpanel / Amplitude → They have built-in cohort + relative date filtering.
SQL → You’d use WHERE cohort_date BETWEEN <dashboard_start> AND <dashboard_end> before grouping by cohort.
👉 So the short answer: Yes, it’s possible—define your cohort on the chosen event date, and tie that to your dashboard’s date filter so cohorts shift relative to the selected period. @https://minecrftgeek.com/