Loading
Marketing Cloud Personalization
Table of Contents
Filter by (0)Add
Select Filters

          No results
          No results
          Here are some search tips

          Check the spelling of your keywords.
          Use more general search terms.
          Select fewer filters to broaden your search.

            Search all of Salesforce Help
            Search all of Salesforce Help
            Create an Einstein Recipe

            You are here:

            1. Salesforce Help
            2. Docs
            3. Marketing Cloud Personalization

            Create an Einstein Recipe

            Einstein Recipes consist of ingredients, exclusions, inclusions, boosters, and variations. Combine these components to suggest content or products according to an individual customer’s behavior and affinities. Personalization uses your recipes to create algorithms that present each customer with their own personalized recommendations.

            Required Editions

            Permissions Needed
            To create an Einstein Recipe: A role with Recipes Create/Edit permissions

            After you create an Einstein Recipe, use it in your campaigns to recommend items based on what your customers have previously viewed or purchased. Personalization’s predictive algorithms dynamically adjust to temporary shifts in consumer behavior.

            1. In the Machine Learning section of the main navigation, click Einstein Recipes.
            2. Click New Recipe.
            3. Enter a Name for the recipe.
            4. Click Recommendation Type and select an item type from the dropdown.
            5. On the Ingredients tab, click Add ingredient and select an option from the dropdown.
            6. Select parameters for the ingredient.
            7. Add additional ingredients as necessary.
            8. Click Additional Options:
              1. If your recipe includes multiple ingredients and one of the anchor items isn’t present, you can select Allow missing anchor item for multi-ingredient recipes. This option presents recommendations from ingredients that either don’t require an anchor item or have fulfilled their anchor item requirements.
              2. To include trending items if the recipe doesn’t return any recommendations, select Enable trending product fallback. You can evaluate trending using a maximum setting of Products from the past 30 days.
            9. Click the Exclusions tab.
              1. Click Add exclusion and select an option from the dropdown.
                Note
                Note Options for exclusions and inclusions vary according to the Recommendation Type you select. See Exclusions and Inclusions in Einstein Recipes.
              2. To use the selected option as an inclusion, click Include.
              3. Select parameters for the exclusion or inclusion.
                Note
                Note If you select the Enable trending product fallback option for an ingredient or for the campaign, Personalization can recommend an exclusion if that excluded item is trending.
              4. Add additional exclusions or inclusions.
            10. Click the Boosters tab.
              1. Click Add booster and select an option from the dropdown.
                Note
                Note Options for boosters vary according to the Recommendation Type you select. See Boosters in Einstein Recipes.
              2. Select parameters for the booster.
              3. Add additional boosters.
            11. Click the Variations tab.
              1. Click Add variation and select an option from the dropdown.
                Note
                Note Options for variations vary according to the Recommendation Type you select. See Variations in Einstein Recipes.
              2. Select parameters for the variation.
              3. Add additional variations as necessary.
                Note
                Note You can add only one dimensional and one randomized variation per recipe.
            12. Save your work.
              Note
              Note Your recipe must have a unique name and at least one ingredient before you can save it.
            13. To compile the algorithms that Personalization uses to present recommendations, click Train.
              Each night, Personalization automatically retrains live recipes that are actively in use. Active recipes are ones that Personalization has used in the past week. Depending on your dataset and customer base, training can take several hours.
            14. When you’re finished creating and training a recipe, be sure to test it before publishing it.
             
            Loading
            Salesforce Help | Article