Einstein Recommendation Builder

Note You can provide feedback and suggestions for Feature in the Salesforce Einstein group in the Trailblazer Community.

Unlike out-of-the box recommendation solutions, Einstein Recommendation Builder lets you create custom recommendations based on standard or custom Salesforce objects.

Einstein Recommendation Builder helps you intelligently recommend records from one Salesforce object to records from a related object. Intelligent recommendations create opportunities to:

• Improve customer satisfaction and retention.
• Grow your business by way of relevant offerings.
• Benefit from operational efficiencies through better matching between business needs and resources.

Owing to its flexible configuration options, Einstein Recommendation Builder provides intelligent recommendations in various contexts across business functions and industry verticals. Here are some examples:

• Cross-Sell Recommendation: Recommend additional products and services to existing customers.
• Next Best Action: Recommend the best action or solution for a customer issue.
• Next Best Offer: Recommend the most relevant offer or promotion to customers and prospects.
• Work Order Enrichment: Recommend relevant parts to carry for service visits by field technicians.
• Job Candidate Recommendation: Recommend relevant candidates who are most likely to accept a job, or vice versa.
• Partner Solution Recommendation: Recommend the most relevant partner solutions to your customers.

• Considerations for Setting Up Einstein Recommendation Builder
  Before you start building recommendations, here are a few things to keep in mind.
• Upgrade to the Paid Version of Einstein Recommendation Builder
  If you use the Try Einstein version of Einstein Recommendation Builder, you can upgrade to a paid license to build recommendations without limits when you're ready.
• Set Up Einstein Recommendation Builder
  Turn on Einstein Recommendation Builder to build and deliver personalized recommendations. Or turn it off if you don’t want to build or make recommendations right now.
• Recommendation Data Terminology
  When using Einstein Recommendation Builder, it’s helpful to understand some terms we use for sets of data.
• Object Type Examples
  You use three types of recommendation objects to build recommendations.
• Plan Your Recommendation
  Before you start building a recommendation, consider what you want to recommend and who or what you want to recommend it to. Then, determine how to translate these elements to Salesforce objects and their relationships.
• Build a Recommendation
  As with any AI solution, building a good recommendation is an iterative process.
• Build a Recommendation Using a Template
  Einstein Recommendation Builder lets you create a custom recommendation or build one from a template. To save time and build a recommendation faster, use a template. As Einstein learns from your data, your recommendations become smarter.
• Refine Your Recommendation
  After creating a recommendation, you can refine it to get optimal results.
• Check Your Data
  Data Checker verifies whether you have sufficient records to build your recommendation. It provides rapid feedback on your recommendation settings and catches issues early. For example, find out sooner during setup if you have enough data or if the same records are used in both positive and negative examples.
• Specify Interaction Examples
  To get the most out of your Einstein recommendation, provide examples of positive and negative interactions.
• Review Your Recommendation’s Scorecard
  After building your recommendation, you can review your recommendation's results in the Einstein Recommendation Builder scorecard. Check out big-picture metrics and get details on the Overview Page.
• Iterate Your Recommendation
  After creating a recommendation, you can iterate it to get optimal results.
• Deploy Your Recommendation
  When your recommendation is ready, deploy it with Einstein Next Best Action.
• Build a Managed Package Recommendation
  A managed package lets independent service vendors (ISVs) and partners easily port custom recommendations across multiple orgs. It also enables faster deployment to customers.