Use Recommenders to Deliver Personalized Experiences

Learning Objectives

After completing this unit, you’ll be able to:

• Explain the power of the cross-channel foundation.
  
• Create a content schema.
  
• Build a Personalization Point with a Decision.
  
• Describe how the same recommender powers web, mobile, email, and batch decisioning use cases.
  

The Cross-Channel Foundation

Congrats! You have built two powerful recommendation strategies. With the recommenders ready to go live, the final step is deployment. This means delivering personalized recommendations to shoppers across every channel such as website, mobile app, and email, without having to rebuild the logic for each one.

This is the core strength of Salesforce Personalization: Build the logic once, deploy it everywhere. The Top Sellers recommender is not limited to the website. That single recommendation logic instantly becomes the universal engine for cross-channel delivery, fueling:

• Web experiences: Drive homepage carousels and product detail page recommendations.
  
• Mobile app experiences: Power native app product tiles and personalized home screens.
  
• Marketing Cloud advanced emails: Inject personalized product recommendations at send time via a repeater component.
  
• Batch decisions for Marketing Cloud Engagement: Generate bulk personalization decisions for Data Cloud segments, then activate them to power outbound campaigns in Marketing Cloud Engagement.
  

In this unit, follow Linda as she packages her recommenders into a single, reusable Personalization Point—the universal decisioning endpoint that delivers 1:1 personalization across the entire Cloud Kicks ecosystem, with zero duplication required.

Design a Content Schema

Before creating a Personalization Point, Linda must determine what data to return in each decision response. A Content Schema acts as the contract between Salesforce Personalization and the delivery channel. To power Cloud Kicks's homepage carousel, Linda defines the specific product fields required to populate each recommended item.

Follow along to create the Content Schema.

1. From the navigation bar, select Content Schemas and select New.
   
2. Select Manual Setup and select Next.
   
3. Add these details.
   
   • Personalization Type: Recommendations
     
   • Content Schema Name: Recs with Header
   • Content Schema API Name is auto populated.
     

4. Select Next.
   
5. Select Add Attribute and enter Attribute Label as Recs Header.
   
6. Select Next.
   
7. In the New Content Schema page, move the following fields from Available DMO Fields to Selected DMO Fields.
   
   • Product Name
     
   • Primary Product Image URL
     
   • Product SKU
     
   • Product Id is automatically selected
     
     
8. Save your changes.
   

With the Content Schema in place, Linda has defined what data gets returned in a recommendations response. Now she’s ready to build the Personalization Point.

Build a Personalization Point

A Personalization Point brings together your recommendations strategy, content schema, and optional targeting rules (via an experiment or set of decisions) into a single object that can be requested across channels for personalized decisions. Follow along as Linda configures the Personalization Point for the Cloud Kicks homepage.

1. In the Personalization app, navigate to Personalization Points and select New.
   
2. Select Manual Setup and select Next.
   
3. Add these details:
   
   • Data Space: default
     
   • Profile Data Graph: Real-Time > RT Profile
     
   • Personalization Point Name: Homepage Recs Banner
     
   • Personalization Point API Name: Auto Populates with Homepage_Recs_Banner
     
   • Personalization Type: Recommendations
     
   • Content Schema: Recs with Header
     
     
4. Save your changes.
   

By selecting the Content Schema, Linda locked in the data contract. Every decision on this Personalization Point will return product recommendations with the exact same fields: Product Name, Image URL, Price, and the custom Recs Header attribute.

Assign a Decision to the Recommender

A Personalization Point needs either an experiment or one or more Decisions to determine which recommender runs and for whom. Follow along to create a simple decision that assigns the Top Sellers recommender as the default for all visitors.

1. In the Personalization app, navigate to Personalization Points, and select Homepage Recs Banner.
   
2. Go to the Related tab.
   
3. In the Personalization Decisions section, select New.
   
4. Add these details and select Next:
   
   • Decision Name: Top Sellers for Everyone
     
   • Decision API Name: Autopopulates with Top_Sellers_for_Everyone
     
5. Add these details for Decision Configuration and select Next:
   
   • Recommender: Cloud Kicks Top Sellers.
     
   • Recs Header: Bestsellers Right Now
6. For Targeting Rules select Always (No Conditions)
   
7. Save your changes.
   

When a shopper visits the Cloud Kicks homepage, the website calls the Homepage_Recs_Banner Personalization Point API. The platform checks for targeting rules, and since there are none, everyone qualifies for the Top Sellers Decision. The Top Sellers recommender then executes, applies its filters, and returns the ranked product list, with the response formatted according to the Recs with Header template. Within this same personalization point, Linda can configure an additional decision targeted at returning visitors and utilize the objective based recommender.

Linda now has a working Personalization Point and is ready to explore how the same endpoint powers other channels before deploying it.

Cross-Channel Application

The Personalization Point you created isn’t limited to a single channel. The same decisioning logic can power experiences across web, mobile, and marketing channels, without rebuilding anything. Here’s how.

Website and Mobile App Experiences

Your website and mobile app can request a decision from the same personalization point. The app sends a server request, gets JSON back, and shows products using the same fields everywhere—for example name, image, price, and link.

There’s no need to re-create the recommendation logic from the website for mobile or vice versa.

Marketing Cloud Advanced Email

You can also use the same recommender from the web/mobile personalization point in email campaigns. In Marketing Cloud Advanced, reference the recommender as a data source on your email and then use merge fields to map the item attributes returned to a component inside a repeater for rendering. At send time, the system makes a server-side API call for each recipient, retrieves their personalized product list, and inserts it into the email.

The same recommender powers web, mobile, and email experiences, without additional setup.

Batch Decisions for Marketing Cloud Engagement

If you’re using Marketing Cloud Engagement, you can use Batch Personalizations. Generate recommendations against the configured personalization point in bulk for a Data 360 segment using the same decisioning logic, then activate those results in your campaigns. This ensures you can still deliver personalized content, even without real-time decisioning.

By reusing a single Personalization Point across channels, you deliver consistent, scalable recommendations wherever your users engage.

Explore It in Action at Cloud Kicks

Now that you’ve learned how a Personalization Point works across channels, let’s bring it back to Cloud Kicks. So far, you’ve built a Top Sellers recommender, defined how the results should be displayed, and exposed it through a Personalization Point, just like Linda.

Cloud Kicks now uses this single setup to power recommendations everywhere. The same Top Sellers carousel appears on the website homepage, surfaces in the mobile app, and is included in the weekly email newsletter. Even outbound campaigns use the same logic through batch processing.

Instead of building separate solutions for each channel, Cloud Kicks reuses one decisioning system to deliver consistent, personalized experiences across every touchpoint.

Next Up

In this unit, you created a Content Schema that defines what product data gets returned, built a Personalization Point with a Decision that assigns the Top Sellers recommender, and explored how it powers experiences across multiple channels. In the next unit, you explore how recommenders drive agentic experiences.

To pass the challenge and move on to the next unit, make sure to select Check Challenge to Earn Points.

Resources

• Salesforce Help: Content Schemas