The Engagement Index is a summary indicator of a known or unknown customer’s historical behavior with your brand. This proprietary metric evaluates a customers’ interactions across your brand’s channels to determine how engaged each customer is with your brand. The Engagement Index metric is derived from the specific marketing dynamics of your brand, and weights each unique action accordingly to produce a customer’s unique Engagement score.

Model Time Windows

  • PREDICTION_WINDOW: Not Applicable

As the Engagement Index is a summary metric of customer behavior, it does not have a PREDICTION_WINDOW. The HISTORICAL_WINDOW is how many days Predictable’s models look back for data. For the Engagement Index model, the default HISTORICAL_WINDOW criteria is that a customer must have clicked an email or triggered a web event within the last 365 days. For email and web events, only data within the HISTORICAL_WINDOW is considered.

Data Processed

  • Email Engagement Data
  • Web / Pixel Engagement Data

This metric does not require transaction data. This is ideal if your company does not have access to transactions or have a proper join key configured.


The model produces a normalized SCORE on a scale of 1.0 to 10.0 that represents a consumer’s engagement with your brand. A customer with a 10.0 is the most engaged. A customer with a 1.0 is the least engaged.

Returned Values:

  • SCORE: the engagement metric
  • CUSTOMER_ID: your unique customer identifiers
  • DATETIME_STAMP: unix timestamp of scoring run
  • MODEL_VERSION: version of platform that scored the run

As this is a metric that Predictable derives from your brand’s unique engagement dynamics and not a predicted score, other returned values are not available.