Personalization Insights Reporting overview

The goal of the Personalization Insights reports is to provide more information on how the Target personalization models behind your AP and AT activities personalize visitor traffic. The Random Forest algorithm is the basis for the Target personalization models.

Because the goal of the Personalization Insights reports is to understand how the Target personalization models decided to send which visitor to what piece(s) of content, the Personalization Insights reports reflect only a sub-segment of all the traffic served by your AP or AT activity. Specifically, the two reports are reflective of all traffic that used the personalization model. In other words, Personalization Insights reports do not consider control traffic or traffic that is served by the overall winner model.

Two Personalization Insights reports are available:

ReportDetails
Automated SegmentsDifferent visitors respond differently to the offers/experiences in your AP/AT activity. This report shows how different automated segments defined by the Target personalization models responded to the offers/experiences in the activity.
Important AttributesIn different activities, different attributes are more, or less, important to how the model decides to personalize. This report shows the top attributes that influenced the model and their relative importance.

Interpreting attributes in Personalization Insights

There are two types of attributes represented in Personalization Insights reports that are used in your AP or Auto Target models:

  • Attributes automatically collected by Target: Target uses a base data set to build its personalization algorithms in AP and AT activities that are reflected in Personalization Insights. See Data Collection for Target’s Personalization Algorithms for data types, example attributes, and their Personalization Insights naming convention. Note that although these attributes are considered, an individual activity’s models might not use all of these attributes in the final model.
  • Attributes passed to Target: See Uploading Data for the Target Personalization Algorithms.

Target provides many ways for you to pass in additional data to Target to enrich the base data set used to build its personalization algorithms in AP and AT activities:

Data TypeDescriptionData Type Naming Convention
Profile Attributes, including profile scripts, Profile Update API, and in-page profile attributesAny information you’ve decided to include in Target’s user profile.
This information could come from profile scripts, information uploaded using the Profile Update API, or in-mbox profile parameters prefixed with “profile.”
Custom - Profile - [parameter name]
Page Parameters (also called “mbox parameters”)Name/value pairs passed in directly through page code that are not stored in the visitor’s profile for future use.Custom - Mbox Parameter - [parameter name]
Customer AttributesCustomer attributes let you upload visitor profile data via FTP to the Experience Cloud. Once uploaded, leverage the data in Adobe Analytics and Adobe Target.Custom - Customer Attributes - [parameter name]
Shared Audiences (Adobe Audience Manager or Adobe Analytics)Audiences created through Adobe Audience Manager or Adobe Analytics and shared with Target.Custom - Experience Cloud Segment - [segment name]
Shared Audiences (Adobe Experience Platform/Real-time CDP)Audiences created through Adobe Experience Platform/Real-time CDP and shared with Target via Destinations.Custom - Adobe Experience Platform Segment - [segment name]
Shared Attributes (Adobe Experience Platform/Real-time CDP)Attributes created through Adobe Experience Platform/Real-time CDP and shared with Target via Destinations. This feature is currently in Beta.Custom - Adobe Experience Platform Attribute - [attribute name]]
In-Activity Reporting Audiences/ SegmentsAudiences defined in your AP or Auto Target activity during setup in “Goals & Metrics.”Custom - Reporting Segment - [segment name]

Frequently Asked Questions

List of frequently asked questions about Automated Personalization (AP) and Auto-Target Insights reports.

How long does data for Automated Personalization (AP) and Auto-Target models persist?

Automated Personalization (AP) and Auto-Target models are trained on the last 45 days of user behavior (user profiles, impression events, and conversion events) for the activity.

Automated Personalization (AP) and Auto-Target models retain user behavior, training records, and model decision data for 90 days to produce Insights reports. After 90 days, training records and model decisions are discarded. Automated Personalization (AP) and Auto-Target models also retain aggregated experience/offer-level impression and conversion data for reporting purposes for two years. This data is aggregate-level data only and does not contain any individual-level profile data.

Training video: Using the Personalization Insights reports Tutorial badge

Target


Adobe Target Maturity Webinar Series

Adobe Customer Success Webinars

Tuesday, Feb 4, 4:00 PM UTC

Adobe Target innovations, including GenAI, and best practices on AI-powered personalization and experimentation at scale.

Register