Create Recommendations for Einstein Next Best Action
What Is a Recommendation?
Now that you’ve created predictions for your contacts, it’s time to create recommendations.
- A prediction says something about the object in question, for example, “This person probably loves to golf.”
- A recommendation is a suggested action: “Sell this person plaid pants.”
Later on, you create a strategy to link recommendations to the predictions … to give the user a custom best action to take with each customer. Think of a strategy as a filter that sorts through recommendations and assigns them to customers based on predictions made for those customers.
Predictions and recommendations are completely independent of each other. You can create as many recommendations as you like (“Offer discount,” “Send reminder email,” “Sell plaid pants”). Some will apply in some situations and not in others. Having a set of recommendations to draw on gives you more flexibility in coming up with a next best action.
- From the App Launcher (), find and open Recommendations.
If necessary, change the List View from Recently Viewed to All
You should see three recommendations we’ve provided for you.
Fill out the fields as follows.
- Name: Join VeloStars
- Description: Suggest joining VeloStars
The description and name should be long enough and unique enough so that when you refer to these recommendations as building blocks of your strategy, you can use some part of the name/description that’s unique.
- Click Upload Image and then Asset Library.
- Select Join VeloStars Today and click Add.
- Acceptance label: Accept
- Rejection label: Reject
- Action: Join VeloStars Today
Adding an action to a recommendation launches a Lightning Flow action. To save you time, we have pre-loaded flows for this project. For information on how to create flows, see the Resources link.
- Click Save.
Click the Recommendations tab and then the All
Recommendations List View. You should have four recommendations that look