Preact User Health Score

Click on each of the following links to get an idea of how Preact scores are calculated and used together. In this article we will discuss the Preact user health score. 

Please note that this is a summary and human readable explanation of the scientific calculations performed on your data. Each customer is different and may have variations.  

What scores does Preact calculate?

  1. Account Health Score
  2. Account Activity Score
  3. User Health Score
  4. User Activity Score

How do the Preact Health and Activity Scores differ? 

  • The Preact Health Scores answer the questions, "is this user/account using the product as they normally do?" 
  • The Preact Activity Scores answer the question, "is this user/accounts using the product as much as they could be?"

What is the Preact User Health Score?

An output of Preact's data science presented in the form of a score. The score is on a scale of 0-100 and reflects the statistical likeliness that a user will not return to your product based on their behavior trends.
How is the Preact User Health Score displayed?
Like the Account Health Score, User Health Score is displayed on the user page as a number highlighted by color.
Health scores between 50 to 100 are shown in green.
Health scores below 50 display in red.

Why is there a health score on each user?

We believe each user is different and uses your product in a unique way. We also use the behavior patterns of each individual to determine the health of an account. 

How is the user health score calculated?

Preact ingest the in-app usage of each user. We define in-app usage as any click that a user takes in your product. (Learn more here). We then use this data to create a model that determines the statistical likeliness that a user will not continue using your product.  

Example One: Karen logs in and is active every 1-2 days. She has not been active for 21 days and we expected to see her 19 days ago. Her probably of not returning is around 65% -70% which is extremely high; therefore her health score is around 30 and she is in poor health. 

Example Two: Myke logs in and is active every 13 to 14 days. Last time he was active was 13 days ago and we don’t expect to see him again for another 1-2 days. He has a probability of not returning of 0%; therefore his health score is around 100 and he is in good health.  

What if there are behavior pattern changes un-related to churn? 

The Preact health score is adaptive and will normalize over time if it detects a new pattern. For example, if a user starts using your product less frequently because they switched job roles. Preact will adjust to that new pattern over time. 

Where can I view this in Preact? 

A summary view is available by hovering over the seat row on an account page

By clicking into the seat or searching a for a user and pulling up their person page 

What does a health score of TBD mean? 

TBD stands for "To Be Determined." This means we have not collected enough data on how a user or account is using your product to assign a Preact score. 

How is the User Health score used in the Preact Account Health Score?

The user health scores are rolled up and apart of an overall account score. Learn more here



Have more questions? Submit a request


Powered by Zendesk