You are here:
Einstein and Data Usage in Marketing Cloud Engagement
Einstein is built on data. When using Marketing Cloud Engagement, review this table to understand the type of data used by Einstein features.
The table lists both the Customer Data that is used, which is data submitted by the Customer to our services as defined in our Main Services Agreement (MSA), and the usage data that is used, which is data relating to users’ interactions with Salesforce. Data submitted to an Einstein feature may be used to train AI models, to improve your services and features, or to develop new features that you can access without additional cost.
The table also indicates which features use global models, which are models that look for aggregated trends across multiple Salesforce Customers. The Customer has control over whether their data contributes to global models. For more information on how to control how your data is used, see Global Model Data Opt-In and Opt-Out for Marketing Cloud Engagement.
| Cloud or Package | Feature | Data Used | Generative AI Model Used | Global Model Used |
|---|---|---|---|---|
| Einstein available in Marketing Cloud Engagement Corporate, Enterprise, and Enterprise + editions. | Subject Line Generation in Einstein Copy Insights and Content Builder | Admin submitted brand identity and brand personalities. Marketing submitted key message. | Yes - OpenAI | No |
| Body Copy Generation in Einstein Copy Insights and Content Builder | Admin submitted brand identity and brand personalities. Marketing submitted key message. | Yes - OpenAI | No | |
| Send Time Optimization for Email | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), Customer sending pattern data and metadata, Email address (anonymized when using global models | No | Yes - optional | |
| Send Time Optimization for Mobile | App Engagement (Push, TapOpen, AppOpen, TimeInApp), Customer sending pattern data and metadata | No | No | |
| Engagement Scoring for Email | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), Customer sending pattern data and metadata, User Retention and Subscription History, Subscriber web interaction events (view, search, profile, and purchase), customer-defined categorization thresholds, Email address (anonymized when using global models) | No | Yes - optional | |
| Engagement Scoring for Mobile | App Engagement (Push, TapOpen, AppOpen, TimeInApp), Customer sending pattern data and metadata, customer-defined categorization thresholds |
No | Yes - optional | |
| Engagement Frequency | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), Customer sending pattern data and metadata, Subscriber ID | No | No | |
| Product Recommendations (Email) | Behavioral Data (pageviews, carts, conversions, categories, wishlists), recommendation engagement (recommendation views, clicks, carts, purchases), and User Attributes provided through Collect.js | No | No | |
| Product Recommendations (Web) | Behavioral Data (pageviews, carts, conversions, categories, wishlists), recommendation engagement (recommendation views, clicks, carts, purchases), and user attributes provided through Collect.js | No | No | |
| Content Selection | Email Engagement (email sends, opens/selections, clicks), optional start-end dates, optional simple rules that exclude assets based on meta-data or previous selections for subscribers. | No | No | |
| Content Testing | Email Engagement (email sends, opens, clicks) | No | No | |
| Content Tagging | Images | No | No | |
| Subject Line Tester | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), subject lines | No | Yes - required | |
| Copy Insights | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), subject lines | No | No | |
| Messaging Insights | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), Customer sending pattern and campaign data and metadata (for example, batch sends, Journey Builder sends, Automation Studio sends, transactional sends, triggered sends, past, and future automations), subject lines | No | No | |
| Einstein Metrics Guard | Email Engagement (email sends, opens, clicks, unsubscribes, bounces, spam complaints), Customer sending pattern data and metadata | No | No |
