what is the use case when using in the retention report the break down by time? why if I do it by month is different than if I expand the wighted average without the breakdown by time in month?
Hi Juan I., 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
To answer your question about using the breakdown by time in the Retention report, let's refer to the relevant information from Mixpanel's documentation.Breaking down by time in the Retention report allows you to analyze retention patterns across different time periods. According to the Mixpanel documentation on Advanced Filter & Breakdown Usage, you can break down your results by the "Time" event property, which includes options like Hour, Day, Week, Month, Quarter, Year, Hour of Day, or Day of Week.The difference you're noticing between breaking down by month and expanding the weighted average without the breakdown is likely due to how Mixpanel calculates retention for segmented (breakdown) and unsegmented queries.As explained in the Retention FAQ: 1. For unsegmented queries (without breakdown), Mixpanel uses an intervalized average method. It calculates retention for each cohort based on the selected birth interval unit (day/week/month) and then summarizes it into one line by taking the average of all complete buckets.2. For segmented queries (with breakdown), Mixpanel uses the same intervalized average method, but for each particular property value or cohort segment. This means that when you break down by month, you're seeing the retention calculated separately for each month's cohort.The main advantages of this approach are: 1. It doesn't include users in incomplete periods, giving all users an equal chance to qualify for later retention buckets. 2. It allows for more granular analysis, as you can see how individual birth interval cohorts are performing within a particular segment.This difference in calculation methods explains why you might see different results when breaking down by month compared to expanding the weighted average without a breakdown.Remember that the weighted average is calculated using a specific formula that takes into account the number of users who enter each cohort, as detailed in the FAQ section of the documentation.
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! 📚
Certainly! I'd be happy to provide an example to illustrate the difference between breaking down by time and expanding the weighted average in the Retention report. However, the knowledge sources provided don't contain a specific example for this scenario. To give you a proper example, I would need access to actual Mixpanel data or a more detailed explanation from their documentation. Without this information, I can't provide a concrete, accurate example that would truly demonstrate the difference you're asking about.What I can do is explain the concept a bit further based on the information we have: When you break down by month, you're essentially creating separate retention analyses for each month's cohort. This allows you to see how retention varies from month to month. For instance, you might see that users who joined in January have different retention patterns compared to those who joined in June.On the other hand, when you expand the weighted average without a breakdown, you're seeing an overall picture of retention across all time periods, where each period is weighted based on the number of users who entered during that time.The Retention FAQ provides a sample calculation for the weighted average, but it doesn't give a full example comparing this to a breakdown by time.If you need a specific example with real numbers, I recommend reaching out to Mixpanel's support team or checking if they have any case studies or more detailed tutorials in their documentation that aren't included in the knowledge sources provided to me.
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 this is the data for us $time,Date,Total Users,< 1 Week,Week 1,Week 2,Week 3,Week 4,Week 5,Week 6,Week 7,Week 8,Week 9,Week 10,Week 11,Week 12 $overall,$average,37.5,6.2975,2.3673,2.2473,1.9551,2.7095,3.1044,3.5364,3.0,3.4059,4.0933,3.0,2.4533,2.0 $overall,2024-12-09,34,5,4,1,2,3,4,3,3,3,3,3,3,2 $overall,2024-12-16,41,5,2,3,3,4,3,4,3,4,5,4,2, $overall,2024-12-23,26,6,2,2,2,2,2,3,3,3,2,1,, $overall,2024-12-30,30,5,3,2,2,3,3,5,3,5,4,,, $overall,2025-01-06,20,7,2,3,2,2,2,2,3,3,,,, $overall,2025-01-13,31,10,3,4,1,3,4,1,2,,,,, $overall,2025-01-20,28,5,2,0,0,1,1,0,,,,,, $overall,2025-01-27,35,4,3,4,3,3,1,,,,,,, $overall,2025-02-03,34,5,2,1,2,2,,,,,,,, $overall,2025-02-10,45,7,1,0,2,,,,,,,,, $overall,2025-02-17,39,10,2,1,,,,,,,,,, $overall,2025-02-24,54,12,0,,,,,,,,,,, $overall,2025-03-03,43,10,,,,,,,,,,,, Dec 2024,$average,5.6,5.0,1.5,0.3,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0 Dec 2024,2024-12-09,6,5,4,1,0,0,0,0,0,0,0,0,0,0 Dec 2024,2024-12-16,6,5,1,0,0,0,0,0,0,0,0,0,0, Dec 2024,2024-12-23,6,6,0,0,0,0,0,0,0,0,0,0,, Dec 2024,2024-12-30,2,2,0,0,0,0,0,0,0,0,0,,, Dec 2024,2025-01-06,0,0,0,0,0,0,0,0,0,0,,,, Dec 2024,2025-01-13,0,0,0,0,0,0,0,0,0,,,,, Dec 2024,2025-01-20,0,0,0,0,0,0,0,0,,,,,, Dec 2024,2025-01-27,0,0,0,0,0,0,0,,,,,,, Dec 2024,2025-02-03,0,0,0,0,0,0,,,,,,,, Dec 2024,2025-02-10,0,0,0,0,0,,,,,,,,, Dec 2024,2025-02-17,0,0,0,0,,,,,,,,,, Dec 2024,2025-02-24,0,0,0,,,,,,,,,,, Dec 2024,2025-03-03,0,0,,,,,,,,,,,, Feb 2025,$average,6.6765,3.1724,0.7917,0.878,0.7429,1.1667,1.7857,2.7391,2.75,3.4286,4.0909,3.0,1.3636,0.0 Feb 2025,2024-12-09,5,0,0,0,0,0,0,0,2,3,3,3,3,0 Feb 2025,2024-12-16,6,0,0,0,0,0,0,3,3,4,5,2,0, Feb 2025,2024-12-23,3,0,0,0,0,0,2,3,3,3,2,0,, Feb 2025,2024-12-30,6,0,0,0,0,2,3,5,3,3,0,,, Feb 2025,2025-01-06,3,0,0,0,2,2,2,2,3,0,,,, Feb 2025,2025-01-13,5,0,0,2,1,3,4,0,0,,,,, Feb 2025,2025-01-20,2,0,2,0,0,1,1,0,,,,,, Feb 2025,2025-01-27,5,1,3,4,3,2,0,,,,,,, Feb 2025,2025-02-03,6,5,2,1,2,0,,,,,,,, Feb 2025,2025-02-10,7,7,1,0,0,,,,,,,,, Feb 2025,2025-02-17,10,10,0,0,,,,,,,,,, Feb 2025,2025-02-24,10,10,0,,,,,,,,,,, Feb 2025,2025-03-03,0,0,,,,,,,,,,,, Jan 2025,$average,6.7778,4.9556,1.7111,2.0222,1.2222,1.2195,0.9167,1.08,0.6667,0.0,0.0,0.0,0.0,0.0 Jan 2025,2024-12-09,4,0,0,1,2,3,4,3,3,0,0,0,0,0 Jan 2025,2024-12-16,5,0,1,3,3,4,3,3,0,0,0,0,0, Jan 2025,2024-12-23,2,1,2,2,2,2,1,0,0,0,0,0,, Jan 2025,2024-12-30,7,3,3,2,2,2,0,0,0,0,0,,, Jan 2025,2025-01-06,7,7,2,3,2,0,0,0,0,0,,,, Jan 2025,2025-01-13,11,10,3,3,0,0,0,0,0,,,,, Jan 2025,2025-01-20,5,5,0,0,0,0,0,0,,,,,, Jan 2025,2025-01-27,4,4,0,0,0,0,0,,,,,,, Jan 2025,2025-02-03,0,0,0,0,0,0,,,,,,,, Jan 2025,2025-02-10,0,0,0,0,0,,,,,,,,, Jan 2025,2025-02-17,0,0,0,0,,,,,,,,,, Jan 2025,2025-02-24,0,0,0,,,,,,,,,,, Jan 2025,2025-03-03,0,0,,,,,,,,,,,, Mar 2025,$average,4.7895,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.6667,2.0 Mar 2025,2024-12-09,2,0,0,0,0,0,0,0,0,0,0,0,1,2 Mar 2025,2024-12-16,4,0,0,0,0,0,0,0,0,0,0,3,2, Mar 2025,2024-12-23,2,0,0,0,0,0,0,0,0,0,2,1,, Mar 2025,2024-12-30,5,0,0,0,0,0,0,0,0,4,4,,, Mar 2025,2025-01-06,3,0,0,0,0,0,0,0,2,3,,,, Mar 2025,2025-01-13,2,0,0,0,0,0,0,1,2,,,,, Mar 2025,2025-01-20,0,0,0,0,0,0,0,0,,,,,, Mar 2025,2025-01-27,2,0,0,0,0,1,1,,,,,,, Mar 2025,2025-02-03,2,0,0,0,0,2,,,,,,,, Mar 2025,2025-02-10,2,0,0,0,2,,,,,,,,, Mar 2025,2025-02-17,2,0,2,1,,,,,,,,,, Mar 2025,2025-02-24,2,2,0,,,,,,,,,,, Mar 2025,2025-03-03,10,10,,,,,,,,,,,, undefined,$average,28.7485,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0 undefined,2024-12-09,25,0,0,0,0,0,0,0,0,0,0,0,0,0 undefined,2024-12-16,32,0,0,0,0,0,0,0,0,0,0,0,0, undefined,2024-12-23,18,0,0,0,0,0,0,0,0,0,0,0,, undefined,2024-12-30,19,0,0,0,0,0,0,0,0,0,0,,, undefined,2025-01-06,13,0,0,0,0,0,0,0,0,0,,,, undefined,2025-01-13,18,0,0,0,0,0,0,0,0,,,,, undefined,2025-01-20,22,0,0,0,0,0,0,0,,,,,, undefined,2025-01-27,27,0,0,0,0,0,0,,,,,,, undefined,2025-02-03,27,0,0,0,0,0,,,,,,,, undefined,2025-02-10,37,0,0,0,0,,,,,,,,, undefined,2025-02-17,29,0,0,0,,,,,,,,,, undefined,2025-02-24,42,0,0,,,,,,,,,,, undefined,2025-03-03,33,0,,,,,,,,,,,,
Let me check the knowledge base for you again. One moment! 📚
Thank you for providing this data. It appears to be a detailed retention report from Mixpanel, broken down by time (months) and showing retention rates for different cohorts over several weeks. This example perfectly illustrates the difference between breaking down by time and the overall weighted average in a Retention report.Let's analyze this data to understand the differences: 1. Overall Retention ($overall): The $average row shows the overall weighted average retention across all cohorts. For example, the Week 1 retention rate is 2.3673%.2. Breakdown by Month: The data is broken down into months (Dec 2024, Jan 2025, Feb 2025, Mar 2025), each with its own $average row.3. Differences in Retention Rates: - Dec 2024: Week 1 retention is 1.5% - Jan 2025: Week 1 retention is 1.7111% - Feb 2025: Week 1 retention is 0.7917% - Mar 2025: Week 1 retention is 0% (likely due to incomplete data)The overall Week 1 retention (2.3673%) is different from each month's retention rate because it's a weighted average of all cohorts, taking into account the number of users in each cohort.This breakdown allows you to see how retention varies by month. For instance, you can observe that January 2025 had better Week 1 retention (1.7111%) compared to December 2024 (1.5%) or February 2025 (0.7917%).The benefit of this breakdown is that it allows you to identify trends or issues specific to certain time periods. For example, you might investigate why February 2025 had lower Week 1 retention compared to other months.On the other hand, the overall weighted average gives you a big-picture view of your retention across all time periods, which is useful for tracking overall performance.This example demonstrates how breaking down by time in the Retention report can provide more granular insights into your user retention patterns, allowing you to spot trends, seasonality, or issues that might be masked in the overall average.
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.
