Federated Audience Composition (FAC) Use Case Scenarios

8 min

Product documentations & Availability

Federated Audience Composition (FAC) in Experience Platform enables you to access and create audiences with corresponding high-value attributes from enterprise data warehouses, which enrich and supplement the real-time customer profile and audiences in Experience Platform for improved segmentation, targeting, activation, and delivery of impactful customer experiences. Using Federated Audience Composition, a virtual database is created by linking remote databases through metadata. This approach simplifies access, reduces duplication, and enhances end-user experience.

Federated Audience Composition (FAC) Overview

Federated Audience Composition (FAC) Tutorial

Getting started with audience composition

Federated Audience Composition FAQ

Federated Audience Composition feature is enabled at sandbox level.

Enabling this feature creates product profile, of the format ACP_FAC - <sandbox> - admin, in Adobe Experience Platform product. Users need to be added to this product profile, to be able to access feature.

Healthcare scenario

Database connection

Name: Healthcare Datasource

Type: Snowflake

Entities & Relationships

📄 Healthcare - Data Model

Healthcare Data Model

Compositions

Use-case 1: Healthcare - Prescription reminder

Segment customers based on prescription due in few days or past due and provided consent to share data for personalisation.

Create audience:

Target profiles, who have made a purchase in last 3 months and marketing consent to "Share" data for personalization.
Enrich filtered audience with personalize attributes, such as, claims in last few months, offers redeemed, translate refill reminder (simple expression), number of checkups (aggregated value).
Deduplicate target audience by prescription refill due days (i.e. refill due recent)
Split the audience:
- Refill due: REFILL_DUE_IN_DAYS greater than 0
- Refill past due: REFILL_DUE_IN_DAYS less than 0
Result audiences: Healthcare - Prescription due for refill, Healthcare - Prescription past due
Identity: Email

Healthcare prescription reminder

Use-case 2: Healthcare - Scorebased segment

Create audience:

Segment customers based on healthcare industry related scores and provided consent to share data for personalisation.

Target profiles with scores related cardiac care, critical illness risk, propensity for healthcare family plans and marketing consent to "Share" data for personalization.
Enrich filtered audience with personalize attributes, such as, claims in last few months, offers redeemed, number of prescriptions in last 3 months (aggregated value).
Split the audience:
- Cardiac care: CARDIAC_CARE_PRODUCT_PROPENSITY greater than or equal to 80
- Critical illness care: CRITICAL_ILLNESS_RISK greater than or equal to 40
- Family plan prospects: HEALTHCARE_FAMILY_PLAN_PROPENSITY greater than or equal to 50
Result audiences: Healthcare - Cardiac care patients, Healthcare - Critical care patients, Healthcare - Family plan prospects
Identity: Email

Use-case 3: Healthcare - Critical Illness Care

Critical care audiences

Segment customers based on critical illnesses and clinical services opted by targets, along with consent to share data for personalization

Create audience:

Target profiles with health records in last 3 months, sponsered healthcare plan, identified with illness (categoriesed as critical diseases) and marketing consent to "Share" data for personalization. ( REPORT_DATE on or after 3 months ago ) AND ( CRITICAL equal to Yes ) AND ( HEALTHCARE_SPONSERED is true ) AND ( SHARE is true )
Change the resultant target dimension to match with Profile schema.
Split the audience:
- Cardiac care: CARDIAC_CARE_TYPE included in Cardiology, Pulmonology
- Oncology care: CARDIAC_CARE_TYPE included in Oncology, Emergency Medicine, Endocrinology
- Other critical diseases (Type: Complement)
Perform data reconciliation, enrichment and deduplication activities to filter target and personalization attributes
Result audiences: Healthcare - Cardiac patients prescription, Healthcare - Oncology patient care, Healthcare - Recommendation for clinical trial patients
Identity: Email

Destination

While activating audience and enriched attributes, you could leverage existing "Federated Audience Composition" (Type: Adobe Applications) destination for Demo System Next, which is a Snowflake database; or, you may choose to create a new destination.

Use the new FAC destination icon, with Adobe logo.

While creating federated database connection in Experience Platform, it is a good practice to refer "Working schema" property (Working schema is the schema in the federated database, with required privilege for the connection user to create/manage tables).

By specifying a "Working schema" property for the federated database connection in Experience Platform, you're ensuring a cleaner and more organized database structure. This allows:

Separation of Concerns: Activation data tables will be created in a dedicated schema, keeping them separate from the original read schema.
Cleaner Table Names: The read schema will not get cluttered with the alphanumeric table names generated by the activation service, which helps maintain clarity and ease of navigation.
Controlled Privileges: The working schema will have the required privileges for the connection user to create and manage tables, ensuring proper security and access control.

This approach enhances the maintainability and readability of your database, which can be especially important in re-usable environments.

Do not refer the same audience label \ name in "Save segment" activity, in multiple composition workflows.

Reusing the same audience label or name in the "Save segment" activity across multiple composition workflows can lead to conflict / error and data integrity issues. To maintain clarity and avoid any potential conflicts, it's crucial to use unique labels or names for each segment within different workflows. This ensures that each segment is distinctly identifiable and correctly associated with its respective workflow.

Once a "Save segment" activity is executed and the corresponding audience is created, it's essential not to add new attributes (audience mapping properties) or change the data types of existing attributes. This is particularly important because:

Consistency: Changing data types or adding new attributes can lead to inconsistencies in the data, making it challenging to maintain and interpret.
Data Integrity: Ensuring the data remains unchanged preserves the integrity of the audience segment, preventing potential errors or mismatches in subsequent workflows.
Stability: Keeping the structure of the audience data stable avoids disruptions in any downstream processes or integrations that rely on the original schema.

Compatibility: Expressions and data transformations often depend on the original data types. Altering these can result in invalid expressions or incompatible data formats.

(Data type change usually occur in case of expressions.)

When activating an audience using the federated destination category, tables with alphanumeric names are created. To easily identify the target table for a specific segment in the federated destination, it is recommended to include the segment name or a unique label during the data activation process.

Incorporate segment name / unique label (e.g. static string expression like "Prime membership prospect") in "Save audience" activity (Audience mapping) within composition workflow.

Ensure the segment name or unique label is included as an attribute during the "Save audience" activity. This facilitates clear identification and association of the segment.

Audience mapping in "Save audience" activity

If the segment name or unique label is captured as an audience attribute from the composition workflow, it can be added as a "Calculated (mapping) field" during the activation step. This attribute should be included to maintain clarity and ease of identification in the federated destination.

Include unique label at activation

Federated Audience Composition (FAC) Use Case Scenarios

Product documentations & Availability

Healthcare scenario

Database connection

Entities & Relationships

Compositions

Destination

✳️ Best Practices