Personalization Points
A personalization point maps to a part of an experience that is eligible for a decision. It is the value included in a personalization decision request — alongside an individual ID — when a request is made to the Personalization decisioning pipeline.
Examples of personalization points include:
- Homepage hero banner
- Homepage recommendations bar
- Product Detail Page (PDP) recommendations bar
- Pop-up overlay
- Sidebar article suggestions
Creating a Personalization Point
Navigate to the Personalization app → Personalization Points → New Personalization Point.
Configuration Options
| Field | Description |
|---|---|
| Name | A unique, descriptive name for the point (e.g., home_hero, pdp_recommendations) |
| Data Space | Everything in Data Cloud is data-space scoped. The selected data space determines which profile data graphs and response templates are available. |
| Data Graph | The profile data graph to use for this point. This impacts: (1) which recommenders are available on decisions, (2) which data points are available for targeting rules. |
| Personalization Type | Either Manual Content or Recommendations (see Personalization Types) |
| Response Template | Defines the shape of the decision response. Only templates matching the selected data space and personalization type are shown. (See Response Templates) |
| Source | Non-editable field that indicates where the point was created from. For points created via the email builder in Marketing Cloud Growth or Advanced, the source reads "CMS Content". |
| Authentication | Optional. Mark a point as requiring authentication. Requests against this point must use Personalization's authenticated decisioning API with proper credentials. |
Step-by-Step
- Click New Personalization Point
- Enter a name — use a machine-friendly format like
home_recommendations - Select the data space
- Select the profile data graph — this connects the point to the real-time profile data used for targeting and recommender input
- Choose the personalization type — Manual Content or Recommendations
- Select the response template — the template determines what configuration options appear on decisions and what data is returned in responses
- Optionally enable authentication if this point should only be accessed via authenticated API calls
- Save
⚠️ Important: After creating a personalization point, you must configure at least one decision or experiment on it before it can return a response when requested. A point with no decisions or experiments returns a blank response.
How the Profile Data Graph Affects the Point
The profile data graph you select has cascading effects on the entire personalization point:
Recommender Availability
Only recommenders configured with the same profile DG and returning the same DMO as the response template are selectable on decisions.
Targeting Rules
Any data point from the selected profile data graph is available for use in decision targeting logic:
- Direct attributes on the Unified Individual
- Related object attributes (engagements, contact points)
- Calculated Insights
- Segment memberships
Request Behavior
When a personalization request includes multiple personalization points, Personalization assumes that all points in the request use the same profile DG. If one point uses a different DG than a point listed earlier in the request, it will be skipped.
📝 Note: This means you should plan your personalization points so that points requested together share the same profile data graph.
Reusability Across Channels
A personalization point is reusable across channels. Any channel that can call the Personalization decisioning API can request a decision for any personalization point. This includes:
- Web — via the Salesforce Interactions SDK (
SalesforceInteractions.Personalization.fetch()) - Mobile — via the Mobile SDK or direct API call
- Email — via Marketing Cloud Growth / Advanced email builder
- Server-side — via direct REST API calls
- Agentforce — via Invocable Actions in Flow
Channel-specific experience wrappers and templates (like web templates or mobile UI components) control when requests are made, where decision responses are rendered, and how content is displayed. But the underlying personalization point, decisions, and recommenders are shared.
💡 Tip: Design your personalization points around the type of experience (hero banner, recommendations bar, popup), not the specific channel. This maximizes reusability and reduces configuration overhead.
Decision Credits
Each decision that Personalization makes consumes 1 decision credit:
- A request with 1 personalization point = 1 decision credit
- A request with 2 personalization points = 2 decision credits
- A blank response (individual doesn't qualify for any decision) still counts as a decision credit
Plan your personalization point strategy with decision credit consumption in mind.
What Happens Next
Once a personalization point is created, configure one of the following:
- Decisions — Define who gets what content, with targeting rules and priority
- Experiments — A/B test different decisioning strategies on this point
Next: Decisions — Learn how to configure targeting rules and priority to determine who receives which personalized content.