Charting new customer data based on zero trailing three month volume

[a3t]

Hi! Thanks in advance for any help. Hopefully the question and details I've provided are enough, but I'm happy to clarify. 

Here is a sample PBIX file: Drive link 

Goal: To chart the following account data on the y-axis for the new accounts by month (legend=year, x-axis=month):

• Count of accounts
• Volume
• Revenue
• Margin

Parameters:

• "New" is defined as 0 volume in prior 3 calendar months
• An account can be "new" multiple times as shown in test file and snippet below

Example:

• Using the sample data, for 2023, January would show nothing; February would show data on 1 account, March would show combined data for 3 accounts, etc.

Problem:

• April and May data is currently showing in the table to the left, but that's based on multiple measures relative to today.
• I've tried SO MANY different things to get this working properly; it seems so simple in my head, but I'm a novice and can't waste any more time trying to figure it out myself. 
  
  

[amitchandak · Accepted solution]

@a3t , refer if this approach can help

Customer Retention Part 2: Period over Period Retention :https://community.powerbi.com/t5/Community-Blog/Customer-Retention-Part-2-Period-over-Period-Retenti...

[a3t]

Thank you! I was able to rig this to get it to work - still some slight issues but it'll do 🙂