Reply
Highlighted
Frequent Visitor
Posts: 13
Registered: ‎01-25-2018

Returning customers over years

Hello,

I have a set of data with sales from customers. What I want to know is how long do new customers stay at my organisation? So if they buy for the first time in 2013. Do they buy also in 2014? And in 2015? So my solution should look like:

 

YearNew customersActive after 1 yearActive after 2 yearsActive after 3 years
2013100906030
201420019015090
2015150135125100
2016350335296250
2017250224200175
2018150145135120

 

My data is a fact Sales with dim Customers and dim Datetable. In fact sales I have transactiondate, amount, sales and keys to dim tables.

 

Thank you for your help!
Regards,

Jarno Visser

Super User
Posts: 9,477
Registered: ‎07-11-2015

Re: Returning customers over years

Source data would really, really help here. Please see this post regarding How to Get Your Question Answered Quickly: https://community.powerbi.com/t5/Community-Blog/How-to-Get-Your-Question-Answered-Quickly/ba-p/38490

 

But, that being said, you sould be able to use a variation of my Not In Common Quick Measure to find customers in a particular year that are not present in the years before.

 

https://community.powerbi.com/t5/Quick-Measures-Gallery/Not-In-Common/m-p/388273

 

That would solve New Customers. After that, it should be a relatively simple matter to compute the MAX order date for each customer and determine if they are active after a length of time.

 

Very interesting case and would love to spend some cycles on it but too much work to hand enter a bunch of data in the wrong format such that it doesn't translate directly to a solution for you.


Did I answer your question? Mark my post as a solution!

Proud to be a Datanaut!


Frequent Visitor
Posts: 13
Registered: ‎01-25-2018

Re: Returning customers over years

Hi Greg,

 

Here is a file with data from my fact table: https://we.tl/t-jbGwLywdWB

The dim table customers you could create with values(customerid).

 

Hope you can help me!

 

Regards,

Jarno