If you’re a data nerd like me, 🤓  it can be quite exciting to see PX usage data in CS. No longer do CS team have to remain in the dark to find answers to burning questions, like:

• “What are the most popular features?”
• “Who are my customer’s power users?”
• “What engagements are getting the most views?” 

The benefits of gaining access to usage data in CS are endless. But what about NON-usage data? CS teams have often ponder questions like:

• “What users have NOT seen the PX NPS Survey engagement?” CS teams may choose to email NPS surveys via a Journey Orchestrator to those users who have not seen the in-app survey engagement 
• “Who has NOT used any golden features this month?” CSMs could use this information during a customer’s Executive Business Review 

Valuable NON-usage data like this requires some additional configuration to become visible in CS. Let’s consider the use case below to explore how:

Users are onboarded via a series of in-app PX engagements. The CS and Education teams would like to understand the differences in adoption between users who have and have NOT seen the onboarding engagements.

To create a dataset of users who have NOT seen the onboarding engagements, follow these: 

• Use Data Designer to create a new Object from which to source info on users who have NOT seen the engagements. 
• Create 2 datasets in Data Designer:
  • Dataset #1- “Active Users”: From Person Time Series Daily, pull in all users who have been active in your application within a certain time period (90 days for example). Here, we are fetching all users where the "User Event Type = Session_Initailized" and "User Event Count is greater than or equal to 1" within the last 90 days.
  • Note: Add additional filtering criteria to the "Active Users" dataset to ensure we are only pulling in users that qualify to see the engagements of interest. We don't want to get false positives where we have a whole list of users who haven't seen the engagement because they didn't qualify to get the engagement in the first place.
  • Dataset #2- “Engagements Seen”: From Person Time Series Daily, fetch the list of users who have seen the engagement(s) using these filters: "Engagement Name = Engagement(s) of interest" and "Engagement Events Count is greater than or equal to 1" 
  • 
• Merge the Datasets: Retain the list of all active users (keep all records on the "left" if that's the side where your Active Last 90 Days list is), and add in the Engagement Count from Dataset #2 on the merge. 
• 
• Now when we query the data from the new Object, we can filter for users where the Engagement Count is NULL- this will give us the list of users who haven't seen the engagement(s)

🪄 Voila! By creating a new data design that merges together all eligible active users with usage data, we now have a dataset from which to source NON-usage data. Be sure to schedule the data design to run on a routine basis to ensure that the data remains up to date and the dependencies built upon this dataset are accurate.

Happy building!