Thomas Francis (ThomasF3903)
Technology Consulting Manager
ThomasF3903 Member since 2014 1 post
Posted: May 11, 2020
Last activity: May 12, 2020
Posted: 11 May 2020 6:46 EDT
Last activity: 12 May 2020 9:30 EDT

How to implement Delayed Learning on ADM to tag a positive outcome basis revenue uplift


We are working to design and deploy ADM on the NBA portal of our client. The client's expectations are that the NBA ADM engine would not only drive success rates but prioritize value/revenue uplift over success rate. I read somewhere that one of the companies in the US is using the delayed learning in PEGA to embed the value uplift aspect into the model learning.


1. Store the  decision data for a certain period (example 10 days)

2. Even if the customers accept on 2nd day, it is parked and not fed into the ADM

3. Define the outcome after 10 days basis the below condition: If response='Accept' and Rev After 10 days  > Rev before 10 days  then 'POSITIVE' else 'NEGATIVE'

I need your valuable input on the same. How exactly can we apply a similar concept in our design.

Hoping for a solution.




***Edited by Moderator Marissa to update Pega Academy to General; Content Type from Discussion to Question***


Decision Management
Moderation Team has archived post, This thread is closed to future replies. Content and links will no longer be updated. If you have the same/similar Question, please write a new Question.