Personalization Points

This guide walks you through building manual content and recommendations personalization points. You will configure personalization decisions, set up targeting rules, create an experiment with cohorts, and simulate personalization requests using Postman.

[ACTION REQUIRED: Review Note] - NOTE: Before starting the hands-on exercises in this workshop, watch the video to learn more about personalization points and decisions or experiments that you can configure within them.

Build a Manual Content Personalization Point

1. Search and select Personalization from App Launcher.
2. Click on the Personalization Points tab.
3. Click New.
4. Click the Manual Setup tile, then click Next.
5. Complete the Personalization Point Properties fields with the following values:
   • Data Space: default
   • Profile Data Graph: Real-Time > Profile
   • Personalization Point Name: Homepage Banner
   • Personalization Point API Name: Homepage_Banner
   • Personalization Type: Manual Content
   • Response Template: Banner with CTA

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Personalization Point Properties for Manual Content Personalization Point

6. Click Save.
7. From the list view, open the Homepage Banner personalization point that you just created.

Configure Personalization Decisions

In this exercise, you will create three personalization decisions for your manual content personalization point.

Decision 1: Mountain Bikes

1. Click New on the Personalization Decisions section.
2. Enter the following Decision Properties values:
   • Decision Name: Mountain Bikes
   • Decision API Name: Mountain_Bikes
   • Decision State: Live
3. Click Next.
4. Enter the following Decision Configuration values:
   • CTA: Shop Now
   • CTA Destination URL: /?category=mountain-bikes
   • Header: Hit the Trail
   • Image URL: [ACTION REQUIRED: Update Image Here] - Original Context/URL: agentforce-marketing-9cf347fa7db7.herokuapp.com/img/velo-mountain.jpg
   • Sub-Header: Mountain Bikes for All

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Personalization Attributes in Personalization Decision

5. Click Next.
6. From the Take Action When menu, select All Conditions Are Met.
7. Click Add Group.
8. Select the following values:
   • Resource: Related Attributes > Unified Link Individual > Product Browse Engagement > Product Category
   • Measurement: Count
   • Operator: Is Greater Than
   • Value: 5
   • WHERE: All Conditions Are Met for Product Browse Engagement
   • Operator: Is Equal To
   • Value: Mountain Bikes

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Targeting Rule in Personalization Decision. This targeting rule uses the Product Browse Engagement DMO to identify users who have viewed products in the Mountain Bikes category more than five times.

9. Click Save & New.

Decision 2: Road Bikes

1. Enter the following Decision Properties values:
   • Decision Name: Road Bikes
   • Decision API Name: Road_Bikes
   • Decision State: Live
2. Click Next.
3. Enter the following Decision Configuration values:
   • CTA: Shop Now
   • CTA Destination URL: /?category=road-bikes
   • Header: Hit the Road
   • Image URL: [ACTION REQUIRED: Update Image Here] - Original Context/URL: agentforce-marketing-9cf347fa7db7.herokuapp.com/img/velo-road.jpg
   • Sub-Header: Road Bikes for All

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Personalization Attributes in Personalization Decision

4. Click Next.
5. From the Take Action When menu, select All Conditions Are Met.
6. Click Add Group.
7. Select the following values:
   • Resource: Related Attributes > Unified Link Individual > Product Browse Engagement > Product Category
   • Measurement: Count
   • Operator: Is Greater Than
   • Value: 5
   • WHERE: All Conditions Are Met for Product Browse Engagement
   • Operator: Is Equal To
   • Value: Road Bikes

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Targeting Rule in Personalization Decision

8. Click Save & New.

Decision 3: Generic Hero

1. Enter the following Decision Properties values:
   • Decision Name: Generic Hero
   • Decision API Name: Generic_Hero
   • Decision State: Live
2. Click Next.
3. Enter the following Decision Configuration values:
   • CTA: Shop Sale
   • CTA Destination URL: /
   • Header: Hey [first name], Get Ready for Spring
   • Image URL: [ACTION REQUIRED: Update Image Here] - Original Context/URL: agentforce-marketing-9cf347fa7db7.herokuapp.com/img/velo-generic.jpg
   • Sub-Header: Bikes For All Riders
4. Select the placeholder string [first name] in the Header field.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Select placeholder value in field to replace with merge field

5. Click Add Merge Field next to the Header field.
6. Select the following values:
   • Attribute: Direct Attributes > First Name
   • Merge Field API Name: First_Name
   • Default Text: Rider
7. Click Save.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Targeting Rule in Personalization Decision

8. Click Next.
9. Leave the Take Action When menu set to Always (No Conditions).
10. Click Save.

Simulate Personalization Request

You can simulate a personalization request against your personalization point using Postman.

[ACTION REQUIRED: Review Note] - NOTE: Unauthenticated and Authenticated Personalization Requests. As this personalization point is used on a public-facing website, it operates in an unauthenticated context. For customers who require an additional layer of security, Personalization also supports authenticated decisioning endpoints and personalization point authentication settings.

1. Click Data Cloud Setup.
2. Scroll to the Your Home Org Details section on the Data Cloud Setup page.
3. Select and copy the Tenant Endpoint URL.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Tenant Endpoint in Data Cloud Setup

4. Launch Postman on your computer and sign in to your account.
5. Open the File menu, then select New.
6. Select HTTP.
7. In the URL field, change the HTTP method to POST.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Set HTTP method in Postman

8. In the URL field, paste your Tenant Endpoint and append the path /personalization/decisions to the end of the URL (POST /personalization/decisions).

[ACTION REQUIRED: Update Image Here] - Original Context/URL: URL field in Postman

9. Select the Body tab.
10. Select the raw radio button.
11. Copy the JSON code below and paste into the request body (payload) text area.

{
  "individualId": "d0cf39a5824a008d",
  "personalizationPoints": [
    {
      "name": "Homepage_Banner"
    }
  ]
}

[ACTION REQUIRED: Review Note] - NOTE: Request Payload. In this request payload, the individualId is set to a sample value (d0cf39a5824a008d). In a real implementation, this would represent a device id or user identifier captured from the website. The name corresponds to the API name of the personalization point you created earlier (Homepage_Banner).

Build and Test a Recommendations Personalization Point

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Personalization Point Properties for Recommendations Personalization Point

1. Click Save.
2. From the list view, open the PDP Recs personalization point that you just created.

In this exercise, you will create an experiment to compare the performance of two recommenders.

Configure Experiment

1. Click New on the Experiment section.
2. Enter Recs Test in the Experiment Name field.
3. Click Next.
4. Select [Purchase] Revenue from the Primary Metric menu.
5. In the Secondary Metrics section, select each metric and click the Selections list.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Add metrics to experiment Primary and Secondary Metrics

[ACTION REQUIRED: Review Note] - NOTE: Primary and Secondary Metrics. Primary and secondary metrics are based on engagement signal metrics. The primary metric determines the winning variation. Secondary metrics are tracked for insights but don’t directly decide the outcome.

6. Click Next twice, as no targeting rule is required for this experiment.
7. Expand the Cohort 1 (Control) heading and select the following values:
   • Recommender: Top Sellers
   • Recs Header: Recommended For You
8. Expand the Cohort 2 heading and select the following values:
   • Cohort Percentage: 60
   • Recommender: Max Rev
   • Recs Header: Recommended For You

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Cohorts in an experiment

[ACTION REQUIRED: Review Note] - NOTE: Cohorts. Individuals are randomly assigned to cohorts based on defined traffic percentages. A cohort must include a minimum of 1,000 participants for winner selection. The total percentage of all cohorts must equal 100 percent. Unallocated percentages apply to the control cohort.

9. Click Save.
10. Click Start to start the experiment.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: Start personalization experiment

[ACTION REQUIRED: Review Note] - NOTE: Starting Experiments. Only experiments in the Started state are evaluated at runtime.

Simulate Recommendations Request

You can now simulate a personalization request against the Personalization Experiment using Postman.

1. Open Postman on your computer.
2. Using the same request URL that you used in the previous exercise, update the request body with the required JSON payload.
3. Click Send.
4. View the product recommendations included in the response payload returned by the personalization experiment.

[ACTION REQUIRED: Update Image Here] - Original Context/URL: View Response Payload

Summary

You have successfully built manual content and recommendations personalization points, configured decisions with targeting rules, and deployed A/B testing experiments using cohorts. Finally, you verified the expected behavior by simulating personalization requests via API.