Einstein Engagement Scoring for Email Model Card

Person or Organization

Salesforce Einstein for Marketing Cloud Engagement

Model Date and Version

• October 2021
• Minor changes can occur throughout the release
• Major changes can occur and are communicated via release notes

Model Type

Forecasting, Time-series analysis. Classification, Clustering

General Information

Einstein Email Engagement Scoring predicts five types of engagement scores for each individual consumer contact in each customer’s business unit. The model further classifies each score into one of four categories based on customer-defined thresholds or machine-inferred thresholds. The model analyzes an individual subscriber’s historical events and engagement patterns to forecast future engagement scores. The most recent behavior and engagement patterns are emphasized in the predictions. The web conversion model is a classification model with crafted features based on historical web interaction events. The model clusters predicted engagement scores to derive local threshold boundaries to classify contact engagement levels.

Constraints

A minimum of one email send is required for a contact to have prediction scores.

License

Einstein Email Engagement Scoring is available to Marketing Cloud Engagement customers with any of these editions.

• Corporate Edition
• Enterprise Edition
• Enterprise+ Edition

Primary Intended Uses

The Einstein Email Engagement Scoring model offers five types of application-specific engagement likelihood for each subscriber. This insight gives you a deeper understanding of your subscribers so you can segment further and target better in future campaigns and sends.

Out-of-Scope Use Cases

Anything other than the primary use case is out of scope and not recommended.

Model Input

Einstein Email Engagement Scoring analyzes up to 90 days of historical engagement patterns of the subscribers. The engagement history includes these factors.

• Email sends, bounces, and engagement events including open, click, and unsubscribe, and associated timestamps
• User retention and subscription history
• Subscriber web interaction events including view, search, profile, and purchase, with associated timestamps
• Data and metadata about customer sending patterns, including how campaigns are executed
• Categorization thresholds defined by customers when applicable

The engagement history that Einstein Engagement Scoring for Email analyzes excludes these factors.

• Data purchased or collected from third parties
• Demographic data that is typically stored in SFMC as data extensions or subscriber or contact attributes
• Specific content within the email template or rendered email body

Model Output

• Up to five score types are predicted for each subscriber for each business unit.
• For each score type, the model produces a list of important factors with impact levels.
• For each business unit, the model produces an average score across all subscribers and an associated model confidence level.
• The model produces data extensions that contain each subscriber’s predicted scores and the corresponding category.
• The model recommends a threshold based on your data that you can use to classify different engagement levels for each score type.

Groups

The model doesn’t use demographic data or other data purchased from third-party data providers.

Environment

The model is trained and deployed in the Marketing Cloud Engagement environment.

Model Performance Measures

Model performance metrics include data such as the model fitness score based on the root mean squared error and mean absolute error, and the data richness score.

Customers are responsible for interpreting Einstein’s insights, and they must be aware of any assumptions that they make when acting based on the model’s insights. However, adverse outcomes are possible due to model error.

Scores and Models

Einstein Engagement Scoring scores and models for email are updated weekly. The refresh cadence varies by one to several days based on a customer's individual business unit.