You are here:
Einstein Email and Web Recommendations Model Card
The model in this card analyzes and provides personalized product and content recommendations.
- Model Details
The Einstein Email and Web Recommendations model in Marketing Cloud Engagement personalizes recommendations with training algorithms, parameters, fairness constraints, features, and other applied approaches. - Intended Use
The Marketing Cloud Engagement Einstein Email and Web Recommendations model is intended for these use cases. - Relevant Factors
These factors are associated with the Einstein Email and Web Recommendations model in Marketing Cloud Engagement. - Metrics
Einstein evaluates and monitors model performance metrics to ensure and improve the quality of the model. These performance measures are associated with the Marketing Cloud Engagement Einstein Email and Web Recommendations model. - Training Data
You have a customized version of the model that’s trained on your data alone. Data from one Salesforce customer doesn’t affect the behavior for another Salesforce customer. While model training happens for each customer on their data, the initial development of the model is validated with a representative set of pilot customers’ data. - Ethical Considerations
Review the ethical factors associated with the Marketing Cloud Engagement Einstein Email and Web Recommendations model. To avoid bias and other ethical risks, the Einstein Email and Web Recommendations model doesn’t include demographic data.
Model Details
The Einstein Email and Web Recommendations model in Marketing Cloud Engagement personalizes recommendations with training algorithms, parameters, fairness constraints, features, and other applied approaches.
Intended Use
The Marketing Cloud Engagement Einstein Email and Web Recommendations model is intended for these use cases.
Primary Intended Uses
Einstein Recommendations provides personalized product and content recommendations to help drive revenue on email, web, and other marketing channels including these options.
- People Who Bought This Also Bought
- People Who Bought This Also Viewed
- People Who Viewed This Also Bought
- People Who Viewed This Also Viewed
- User Affinity
- Tag Scenarios
See the full list of Einstein Recommendation Scenarios.
Out-of-Scope Use Cases
Anything other than the primary use case is out of scope and not recommended.
Relevant Factors
These factors are associated with the Einstein Email and Web Recommendations model in Marketing Cloud Engagement.
Model Input
Einstein Recommendations analyzes up to 180 days of historical behavioral data from Collect.js for each business unit. The engagement history includes these factors.
- Pageviews
- Category Views
- Carts
- Conversions
- Wishlists
- Metadata from the product or content catalog is used to build user affinity and provide attribute context. The affinity profile is created from user-tagged metadata from the catalog and is based on behavioral information.
- Output is based on localized behavioral data, catalog content, and user attributes by business unit.
- Business-specific rules added using Rule Manager impact the model
The engagement history that Einstein Metrics Guard analyzes excludes these factors.
- Data purchased or collected from third parties
- Data from other business units
- Demographic Data, which is typically stored in SFMC as data extensions or subscriber or contact attributes
Model Output
- The model output is a list of products or content via email or web recommendation call
- Web recommendation output is a JSON block
- Email Recommendation is a link & image pair
Environment
The model is trained and deployed in the Salesforce Marketing Cloud Engagement environment.
Metrics
Einstein evaluates and monitors model performance metrics to ensure and improve the quality of the model. These performance measures are associated with the Marketing Cloud Engagement Einstein Email and Web Recommendations model.
Model Performance Measures
Model performance metrics include data like model mean absolute error score.
Training Data
You have a customized version of the model that’s trained on your data alone. Data from one Salesforce customer doesn’t affect the behavior for another Salesforce customer. While model training happens for each customer on their data, the initial development of the model is validated with a representative set of pilot customers’ data.
Ethical Considerations
Review the ethical factors associated with the Marketing Cloud Engagement Einstein Email and Web Recommendations model. To avoid bias and other ethical risks, the Einstein Email and Web Recommendations model doesn’t include demographic data.
Consider any assumptions made when deciding on actions based on the model-generated scores that could lead to a potentially adverse outcome.
