Skip to main content
Solved

Divide by Dynamic User Cohort within Custom Formula

  • October 14, 2021
  • 6 replies
  • 971 views

Forum|alt.badge.img

Hi,

I am attempting to find the percentage of users in a specific cohort that performed a few events. So far, I am able to find this percentage by using the Formula feature within the event segmentation chart. I have 42,087 users within this dynamic cohort. To find the percentage, I can take the uniques from each event and divide by the number of users in my cohort. Is there a way to shortcut this so that I can just divide by number of users within the cohort itself? If the cohort grows in size, I would like to have this chart adapt without having to retype the number of users in the cohort for the formula each time. 

Formula I am using 

Instead, could the formula do something like UNIQUES(A) / ”Current Cohort Population” ?

 

Here is my chart for reference:

 https://analytics.amplitude.com/viasat/chart/6bo9mqb?source=redirect%3A+chart+saved

Let me know if you can help out!

-Anna

Best answer by jmagg

Hi @akrause - we have had several customers write in with similar inquiries! At the moment, it is not possible to achieve what you are looking for since we don't have a metric to measure total # of users within a cohort at each data point. We would need to make some sort of Custom Formula along the lines of UNIQUES(A)/COHORTTOTAL(A) - but we don't have a metric for COHORTTOTAL at the moment. Amplitude is an event-based platform so we don't have a way to track users who do not perform events yet still make up a segment. Along the lines of @Saish Redkar’s recommendation, the closest workaround I can think of is to select the event of interest & Any Event on the left module and the desired segment on the right module, then use the custom formula UNIQUES - example setup here. Hope this helps!

View original
Did this topic help you find an answer to your question?

6 replies

Saish Redkar
Expert
Forum|alt.badge.img+10

Hey @akrause 

If I’m understanding your use case correctly, the best approach would be to use the ACTIVE custom formula for this, since you are interested in the percent of active users who performed the event.

Here is an example chart showing the formula usage.

You can also use the “Active %” metric instead of the formula to get a similar result, like this.

Hope this helps!


Forum|alt.badge.img
  • Author
  • New Member
  • 2 replies
  • October 15, 2021

@Saish Redkar,

Thanks for responding! Your response works if I want to filter by Active users. But is there a similar shortcut if I want to divide my uniques by all members of the cohort (both active and inactive)? 


jmagg
Team Member
Forum|alt.badge.img+6
  • Team Member
  • 95 replies
  • Answer
  • October 18, 2021

Hi @akrause - we have had several customers write in with similar inquiries! At the moment, it is not possible to achieve what you are looking for since we don't have a metric to measure total # of users within a cohort at each data point. We would need to make some sort of Custom Formula along the lines of UNIQUES(A)/COHORTTOTAL(A) - but we don't have a metric for COHORTTOTAL at the moment. Amplitude is an event-based platform so we don't have a way to track users who do not perform events yet still make up a segment. Along the lines of @Saish Redkar’s recommendation, the closest workaround I can think of is to select the event of interest & Any Event on the left module and the desired segment on the right module, then use the custom formula UNIQUES - example setup here. Hope this helps!


Forum|alt.badge.img
  • Author
  • New Member
  • 2 replies
  • October 18, 2021

@jmagg ,

Thank you! Is there a way I can send in a recommendation this feature? 


Saish Redkar
Expert
Forum|alt.badge.img+10

@akrause 
You can submit a feature request/idea for this in here.


Hi @jmagg and @jmagg  is there any updates on this feature ? is this now doable? thanks


Reply


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings