Counting Purchases by User Signup Month and Weekly Events
I want to count how many events A (purchases) occurred on a given week when the user signs up within a given month. Can I do it?
I want to count how many events A (purchases) occurred on a given week when the user signs up within a given month. Can I do it?
Hi Luisa S., 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> 👋
Feel free to ask a follow-up question once I've answered by mentioning me with @kapa.ai
Yes, you can achieve this analysis in Mixpanel by combining event filtering and cohorting features: 1. Filter users who signed up within a given month: You can create a cohort of users who performed the "Sign Up" event during your target month. This is done by filtering the "Sign Up" event by date range (e.g., users who signed up between May 1 and May 31).2. Count "Purchase" events for those users in a given week: Once you have your cohort, you can analyze the "Purchase" (event A) activity for this group. In Mixpanel reports (such as Insights), you can filter the "Purchase" event by the cohort you created and specify the week you are interested in.This approach allows you to count how many "Purchase" events occurred in a specific week, but only for users who signed up in a specific month. You can use the date range selector to define both the signup month and the purchase week as needed. The cohort filter ensures that only the relevant users are included in your analysis, and the event filter/count gives you the number of purchases in the desired week Filters Date Range.If you need to automate or further customize this analysis, you could use JQL (JavaScript Query Language) to programmatically filter users by signup date and then count their purchases in the desired week JQL Builtins.If you need step-by-step guidance on setting up the cohort or the exact report, let me know!
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.
Let me check the knowledge base for you again. One moment! 📚
Let me check the knowledge base for you again. One moment! 📚
Yes, you can perform this type of rolling analysis in Mixpanel using cohorts and the Insights report, but there are some nuances to be aware of: 1. Cohort Definition (Rolling Window): - You can define a cohort of users who signed up within a rolling month window. For example, for each week, you can create a cohort of users who signed up in the 30 days (or calendar month) prior to that week. - Mixpanel cohorts are dynamically computed at query time, so if you use a relative date filter (e.g., "signed up in the last 30 days"), the cohort will always reflect the users who meet that criteria as of the time the report is run. This supports your rolling lookback requirement How Cohorts are Computed.2. Counting Events for the Cohort: - In the Insights report, you can filter the "Purchase" event by the cohort you defined. Then, set the time interval to "Week" to see the count of purchases per week for users who signed up in the rolling month window. - As you move from week 44 to 45, the cohort (users who signed up in the last month) will update, and the report will show the number of purchases for the new cohort in each week Using Cohorts in Analysis.3. Rolling Metric Behavior: - This setup ensures that for each week, you are always looking at purchases made by users who signed up in the most recent month (relative to that week). The cohort and the event count both "roll" forward as you move through time.How to set this up: - Create a cohort: Users who did "Sign Up" in the last 30 days (or last calendar month). - In Insights, select the "Purchase" event, filter by your cohort, and set the interval to "Week".Note: If you want to see the cohort size or compare different rolling windows, you can use the cohort as a breakdown or filter in your reports Using Cohorts in Analysis.If you need more granular control (e.g., custom rolling windows or more complex logic), you may need to use JQL or export data for external analysis, but the above approach covers most standard rolling cohort analyses in Mixpanel.Let me know if you need step-by-step instructions for setting up the cohort or the report!
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, you can set up this type of rolling metric in Mixpanel using cohorts and the date range selector, but there are some nuances to be aware of: 1. Cohort Definition (Rolling Month Lookback): - You can create a cohort of users who signed up within the last 30 days (or a rolling month window). This cohort will update dynamically as time moves forward, so when you move from week 44 to 45, the cohort will now include users who signed up in the new rolling month window. This is possible because Mixpanel cohorts are computed dynamically at query time, always reflecting the current definition (e.g., "signed up in the last 30 days") How Cohorts are Computed.2. Counting Events for the Cohort in a Given Week: - In an Insights report, you can filter the "Purchase" event by the cohort you just defined. Then, set the time interval to "Week" and use the date range selector to move your analysis window forward week by week. Each time you shift the week, the cohort and the event count will update to reflect the new rolling month and the purchases in the selected week Using Cohorts in Analysis Date Range.Key Points: - The cohort is always recalculated based on the current time window, so it acts as a rolling filter. - The event count (purchases) is shown for the selected week, filtered to only those users in the rolling cohort.Limitations: - If you want to see this metric as a time series (e.g., a line chart showing each week’s count for the rolling cohort), you can visualize it in Insights by setting the interval to "Week." Each point will represent the number of purchases in that week for users who signed up in the rolling month window ending at that week Visualizing Cohort Size Over Time.If you need to define a more custom rolling window (e.g., not exactly 30 days, or with more complex logic), you may need to use JQL for advanced queries, but the standard cohort and Insights tools cover most rolling cohort use cases.If you need step-by-step instructions for setting this up, let me know!
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.
