What is Einstein Next Best Action in Salesforce

Einstein Next Best Action is a powerful tool that helps guide users by offering personalized recommendations. It works by using flows, strategies, and the Recommendation object to suggest the most relevant actions.

You can show these recommendations on various platforms, like Lightning pages in Salesforce, Experience Cloud sites or even external websites.

When users see a recommendation, they have the option to accept or reject it. Each suggestion usually includes an image, important details like a description and button text, and a flow that runs based on the user’s response.

These recommendations are stored in Salesforce's Recommendation object or can be manually put together when creating a strategy.

Strategies play a crucial role in deciding which recommendations to show, based on your organization's data and processes.

When you place Einstein Next Best Action on a page, you link it to a strategy, determining what recommendations appear for users.

Even with many recommendations, you can easily control what’s displayed in any situation. Strategies can filter recommendations based on available data—whether it’s related to the recommendation itself, the user viewing it, or the record that’s currently displayed.

We will take an example of how the Einstein Next Best Actions work. We will add a Next Best Action component to the Lightning Account record page to offer a gift basket to your accounts.

When a customer accepts the offer, a form will appear to collect the recipient’s name and shipping address. Once the form is submitted, an email request is automatically sent to the shipping department to process the order.

Step 1: Create an Action Flow

• Navigate to Setup -> Flows -> New Flow.
• Select the Flow type as 'Screen Flow' and click the Create button.
• Add a Screen element on the Flow to capture recipient details.
• We will add the 'Name' and 'Address' components to the screen element to capture the name and address of the recipient and assign an API name to each and click Done.

• In the Flow Builder Toolbox, click New Resource with type 'Text Template' to create the email message text that will be sent to the shipping department.

• To create a task for the shipping department, add an Action element to the flow. In the Action dropdown, type Send Email and choose the Send Email action.

• Click Done and Activate the flow.

Step 2: Create a Recommendation

• From the App Launcher (icon), type Recommendations in the Quick Find box and select Recommendations. Then, click New.

• Select the target audiences for the recommendation and click Save.

Step 3: Create a Recommendation Strategy

The recommendation strategy flow controls when and how the recommendation is shown.

• In Setup, type Flows into the Quick Find box, choose Flows, and then click New Flow. Select Use a Template and click Next.
• Choose the Recommendation Strategy flow type and click Create.
• You can see there is a collection variable already created in your flow which is going to return you the collection of recommendations.
• To define which records to use for the recommendation, add a Get Records element to the flow. In the Filter section, set the condition to Id equals recorded and click Done.
• To load potential recommendations into the strategy, add another Get Records element. In the Filter section, set the condition to Name Contains Gift Basket and click Done.

• To make the recommendation output available to Einstein Next Best Action, click + below the Recommendation Assignment element and add an Assignment element.
• For the Value, choose Recommendations from Get Gift Recommendation.
• Finally, save the flow and activate it.

Step 4: Add a Best Next Action component

• Display the Next Best Action recommendation on the Account record page.
• Open an Account record page and from the gear icon click on 'Edit page'.

• Save your changes, then go back to the Account record and refresh the page. You’ll see the recommendation displayed.
• If the account rep clicks Yes, a form will open to enter the name and address. Once the form is completed, an email request is sent to the shipping department to process the order.

Here are a few examples of how you can use Einstein Next Best Action in Salesforce

• Suggest a discount or promotional offer to customers with high engagement scores.
• Propose a personalized upsell or cross-sell offer based on the past purchase history of customers.
• Recommend a follow-up call or meeting when an opportunity reaches a specific stage in the sales pipeline.

Things to Consider While Working with Einstein Next Best Actions

• Strategies require at least one recommendation to function properly.
• Strategies will only load recommendations that are connected to active flows.
• Every Recommendation object must reference a flow; without a flow, recommendations can't be displayed.
• You can create recommendations linked to inactive flows, but they won’t be used by any strategy until the flow is activated.
• Active flows are essential for strategies to properly load and display recommendations.
• To enhance control over recommendations, consider adding a custom category field to the Recommendation object and its layout. This allows better sorting, filtering, and managing of recommendations.

Need Help? Reach me out !!

Do you have a tricky scenario in Apex? Do you need some expert advice or have deep questions you want answers to?

Whether it's a complex problem or guidance on building your flow, book your slot and let's connect on Topmate.

Are you ready to embark on your Salesforce Flow journey? Unleash your potential by diving into our comprehensive Salesforce Flow course.

Whether you're a seasoned pro or just starting, there's something for everyone.

But that's not all. Are you looking for top-notch Salesforce learning resources?

We've got you covered !!

You can supercharge your certification prep with our mock exams series.

Stay ahead, stay informed and start your journey to mastering Salesforce Flows today ♥

We'll be back next Wednesday with fresh insights till that time, continue your exploration and keep learning Geeks 🌐📚