User Cohort Analysis

[Anonymous]

Hi everyone,

I really want to create a cohort analysis for new users in Power BI (like the below example in Excel). So every month we get new users and I want to know how many stay a couple of months later. For example, out of the 38 new users in Jan-20, only 25 users stayed two months later. 

I have a table with all the users for the last 20 months. In this table, I have both Start_date and Last_date. But I need help with the DAX function and I hope someone can help 🙂

[DataZoe]

@Anonymous ,

I have done something similar, in order to see the # people lost and # people gained that make up a month over month change in people. I used this SQLBI example to get mine working:

https://www.daxpatterns.com/new-and-returning-customers/

Hope this helps!

[amitchandak]

@Anonymous , very difficult to suggest without looking at data.

check these

https://finance-bi.com/power-bi-cohort-analysis/
http://funbiworld.com/2019-12-02-cohort-analysis/
https://www.youtube.com/watch?v=tuk1CrigTSI

[Anonymous]

Hi @amitchandak 

I've seen these links, but all of them are on customers, so mine is quite different. I want to look at two date columns instead of looking at if the customer appears again. 

I have a user_table with 4 columns (User_ID, Name, Start_Date and End_Date). If the user is still member than End_Date is null. Start_date is the y-axis as shown in the example from Excel and to the x-axis, I created the below table: