Building Cohorts in Kapa.ai: Steps for Specified Activity Criteria

Jack P. · 2025-08-13T14:00:27.407Z

how do I build the following cohorts? > • Active ≥3 weeks in any 8-week window between Feb–Apr _AND_ inactive ≥4 consecutive months post-April > • Active ≥1 week in ≥8 distinct months in last 12. I have our active event but I can't figure out how to specify events only within specified number of weeks between a period, or the distinct months element.

Building Cohorts in Kapa.ai: Steps for Specified Activity Criteria | Mixpanel Community

2 comments

· Sorted by Oldest

kapa.ai
APP
·
·
Please stand by Jack P. as I search the knowledge base. I normally take about 15 seconds or so. 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
kapa.ai
APP
·
·
Building these advanced cohorts in Mixpanel requires combining event frequency logic with time-bounded and distinct-interval criteria. Here’s what’s possible based on the knowledge sources: ---
1. Active ≥3 weeks in any 8-week window between Feb–Apr AND inactive ≥4 consecutive months post-April
What’s Possible in Mixpanel
- Frequency in a Time Window: You can create a cohort for users who were active (did your “active” event) in at least 3 unique weeks within a custom date range (e.g., Feb–Apr). This is achievable using the cohort builder’s frequency and time filters. - Inactivity After a Period: You can also create a cohort for users who were inactive (did not do the event) for a set period after April.
Limitations
- “Any 8-week window” logic: Mixpanel’s cohort builder allows you to specify activity in a fixed window (e.g., “at least 3 weeks between Feb 1 and Apr 30”), but it does not natively support rolling or sliding windows (e.g., “any 8-week window within Feb–Apr”). You can only specify a fixed date range, not a moving window within that range. - Consecutive Inactivity: You can filter for users who have not performed the event in the last X months, but specifying “≥4 consecutive months post-April” as a strict requirement is not natively supported; you can only approximate this by filtering for no activity in a 4-month window after April.
How to Build (Approximate)
- Step 1: In the Cohort Builder, set: - “Did event [Active] on at least 3 unique weeks between Feb 1 and Apr 30” - Step 2: Add an AND condition: - “Did NOT do event [Active] between May 1 and Aug 31” (for 4 months of inactivity)> Note: This will not capture “any 8-week window” logic, only the fixed Feb–Apr window, and inactivity is for a fixed period, not strictly consecutive months if there’s any activity in between. Creating Cohorts---
2. Active ≥1 week in ≥8 distinct months in last 12
What’s Possible in Mixpanel
- Distinct Months Logic: Mixpanel’s cohort builder allows you to filter users by the number of unique time intervals (e.g., months) in which they performed an event. You can specify “did event in at least 8 unique months in the last 12 months.”
How to Build
- In the Cohort Builder, set: - “Did event [Active] in at least 8 unique months in the last 12 months”This is supported and aligns with Mixpanel’s frequency and time-bucket logic for cohorts Frequency View Creating Cohorts---
Summary
- The second cohort (≥1 week in ≥8 distinct months) is directly supported. - The first cohort can be approximated, but Mixpanel does not natively support “any 8-week window” logic or strict consecutive inactivity periods—only fixed date ranges and frequency within those.If you need more granular or rolling-window logic, you may need to export your data and process it externally before re-importing as a static cohort via CSV Creating a Cohort via a CSV.Let me know if you need step-by-step instructions for setting up these cohorts in the UI!
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.

kapa.ai
APP
·
·
Please stand by Jack P. as I search the knowledge base. I normally take about 15 seconds or so. 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
kapa.ai
APP
·
·
Building these advanced cohorts in Mixpanel requires combining event frequency logic with time-bounded and distinct-interval criteria. Here’s what’s possible based on the knowledge sources: ---
1. Active ≥3 weeks in any 8-week window between Feb–Apr AND inactive ≥4 consecutive months post-April
What’s Possible in Mixpanel
- Frequency in a Time Window: You can create a cohort for users who were active (did your “active” event) in at least 3 unique weeks within a custom date range (e.g., Feb–Apr). This is achievable using the cohort builder’s frequency and time filters. - Inactivity After a Period: You can also create a cohort for users who were inactive (did not do the event) for a set period after April.
Limitations
- “Any 8-week window” logic: Mixpanel’s cohort builder allows you to specify activity in a fixed window (e.g., “at least 3 weeks between Feb 1 and Apr 30”), but it does not natively support rolling or sliding windows (e.g., “any 8-week window within Feb–Apr”). You can only specify a fixed date range, not a moving window within that range. - Consecutive Inactivity: You can filter for users who have not performed the event in the last X months, but specifying “≥4 consecutive months post-April” as a strict requirement is not natively supported; you can only approximate this by filtering for no activity in a 4-month window after April.
How to Build (Approximate)
- Step 1: In the Cohort Builder, set: - “Did event [Active] on at least 3 unique weeks between Feb 1 and Apr 30” - Step 2: Add an AND condition: - “Did NOT do event [Active] between May 1 and Aug 31” (for 4 months of inactivity)> Note: This will not capture “any 8-week window” logic, only the fixed Feb–Apr window, and inactivity is for a fixed period, not strictly consecutive months if there’s any activity in between. Creating Cohorts---
2. Active ≥1 week in ≥8 distinct months in last 12
What’s Possible in Mixpanel
- Distinct Months Logic: Mixpanel’s cohort builder allows you to filter users by the number of unique time intervals (e.g., months) in which they performed an event. You can specify “did event in at least 8 unique months in the last 12 months.”
How to Build
- In the Cohort Builder, set: - “Did event [Active] in at least 8 unique months in the last 12 months”This is supported and aligns with Mixpanel’s frequency and time-bucket logic for cohorts Frequency View Creating Cohorts---
Summary
- The second cohort (≥1 week in ≥8 distinct months) is directly supported. - The first cohort can be approximated, but Mixpanel does not natively support “any 8-week window” logic or strict consecutive inactivity periods—only fixed date ranges and frequency within those.If you need more granular or rolling-window logic, you may need to export your data and process it externally before re-importing as a static cohort via CSV Creating a Cohort via a CSV.Let me know if you need step-by-step instructions for setting up these cohorts in the UI!
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.

Building Cohorts in Kapa.ai: Steps for Specified Activity Criteria

2 comments

1. Active ≥3 weeks in any 8-week window between Feb–Apr AND inactive ≥4 consecutive months post-April

What’s Possible in Mixpanel

Limitations

How to Build (Approximate)

2. Active ≥1 week in ≥8 distinct months in last 12

What’s Possible in Mixpanel

How to Build

Summary

1. Active ≥3 weeks in any 8-week window between Feb–Apr AND inactive ≥4 consecutive months post-April

What’s Possible in Mixpanel

Limitations

How to Build (Approximate)

2. Active ≥1 week in ≥8 distinct months in last 12

What’s Possible in Mixpanel

How to Build

Summary