Adobe Experience Platform Segmentation Service provides a user interface for creating and managing segment definitions.
Working with segment definitions requires an understanding of the various Experience Platform services involved with segmentation. Before reading this user guide, please review the documentation for the following services:
It is also important to know two key terms that are used through this document and understand the difference between them:
In the Experience Platform UI, select Segments in the left navigation to open the Overview tab displaying the Segments dashboard.
If your organization is new to Platform and does not yet have active Profile datasets or merge policies created, the Segments dashboard is not visible. Instead, the Overview tab displays links and documentation to help you get started with segments.
The Segments dashboard outlines key metrics related to your organization’s segment data.
To learn more, visit the segment dashboard guide.
Select the Browse tab to see a list of all the segment definitions for your organization.
This view lists information about the segment definition including the profile count, created date, and last modified date.
You can add additional fields to this display by selecting . These additional fields include breakdown, evaluation method, and job ID.
If breakdown is selected, the display shows a bar graph outlining the percentage of profiles that belong to each of the following calculated profile statuses: Realized, Existing, and Exiting. Additionally, the breakdown shown on the Browse tab is the most accurate breakdown of the segment status. If this number differs with what is stated on the Overview tab, you should use the numbers on the Browse tab as the correct source of information, since the Overview tab numbers only update once per day.
Status | Description |
---|---|
Realized | The count of profiles that qualified for the segment in the last 24 hours since the last batch segment job ran. |
Existing | The count of profiles which remained in the segment in the last 24 hours since the last batch segment job ran. |
Exiting | The count of profiles which exited the segment in the last 24 hours since the last batch segment job ran. |
The evaluation method can either be streaming, batch, or edge. Streaming segments are constantly evaluated as data enters the system. Batch segments are evaluated according to a set schedule. Edge segments are evaluated in real-time, which allow for same page and next page personalization use cases.
On the top of the page are options to add all segments to a schedule and to create a new segment.
Toggling Add all segments to schedule will enable scheduled segmentation. More information on scheduled segmentation can be found in the scheduled segmentation section of this user guide.
Selecting Create segment will take you to the Segment Builder. To learn more about creating segments, please read the section on creating a segment in the user guide.
The right sidebar contains information about all the segments within the organization, listing the total number of segments, the last evaluation date, the next evaluation date, as well as a breakdown of the segments by evaluation method.
Selecting the segment definition’s row provides a summary of the segment definition, including options to either edit or delete the segment, activate the segment to a destination, the qualified audience for the segment, the total audience size, in addition to the segment’s name, description, evaluation method, created date, and last modified date.
You will not be able to delete a segment that is used in a destination activation.
To see more details about a specific segment definition, select a segment’s name within the Browse tab.
The segment details page appears. On the top, there is a summary of the segment definition, information about the qualified audience size, as well as destinations the segment is activated for.
The Segment summary section provides information such as the ID, name, description, and details of the attributes.
Additionally, you are given the option to either activate the segment to a destination or edit the segment. Selecting Activate to destination will let you activate the segment to a destination. For more detailed information on activating a segment to a destination, please read the activation overview.
Selecting Edit segment will bring you to the Segment Builder. For more detailed information about using the Segment Builder workspace, please read the Segment Builder user guide.
The Total audience in segment section shows the total number of profiles that qualify for the segment.
Estimates are generated by using a sample size of that day’s sample data. If there are less than 1 million entities in your profile store, the full data set is used; for between 1 and 20 million entities, 1 million entities are used; and for over 20 million entities, 5% of the total entities are used. More information about generating segment estimates can be found in the estimate generation section of the segment creation tutorial.
The Activated destinations section shows the destinations that this segment is activated for.
Destinations are a feature available with Adobe Real-Time Customer Data Platform, and allow you to export data to external platforms. For more information on destinations, please read the destinations overview. To learn how to activate a segment to a destination, see activation overview.
Underneath is a sampling of profiles that qualify for the segment, detailing information including the Profile ID, first name, last name, and personal email.
The way data sampling gets triggered depends on the method of ingestion.
For batch ingestion, the profile store is automatically scanned every fifteen minutes to see if a new batch was successfully ingested since the last sampling job was run. If that is the case, the profile store is subsequently scanned to see if there’s been at least a 5% change in the number of records. If these conditions are met, a new sampling job is triggered.
For streaming ingestion, the profile store is automatically scanned every hour to see if there’s been at least a 5% change in the number of records. If this condition is met, a new sampling job is triggered.
The sample size of the scan depends on the overall number of entities in your profile store. These sample sizes are represented in the following table:
Entities in profile store | Sample size |
---|---|
Less than 1 million | Full data set |
1 to 20 million | 1 million |
Over 20 million | 5% of total |
More detailed information about each Profile can be seen by selecting the Profile ID. To learn more about a profile’s details, please read the Real-Time Customer Profile user guide.
Selecting Create segment in the top-right corner opens the Segment Builder workspace, where you can begin creating a segment definition.
Segment Builder provides a rich workspace that allows you to interact with Profile data elements. The workspace provides intuitive controls for building and editing rules, such as drag-and-drop tiles used to represent data properties.
For more detailed information about using the Segment Builder workspace, please read the Segment Builder user guide.
Once segment definitions have been created, you can then evaluate them through on-demand or scheduled (continuous) evaluation. Evaluation means moving Real-Time Customer Profile data through segment definitions in order to produce corresponding audiences. Once created, the audiences are saved and stored so that they can be exported using Experience Platform APIs.
On-demand evaluation involves using the API to perform evaluation and build audiences as needed, whereas scheduled evaluation (also known as ‘scheduled segmentation’) allows you to create a recurring schedule to evaluate segment definitions at a specific time (at a maximum, once daily).
Enabling your segment definitions for scheduled evaluation can be done using the UI or the API. In the UI, return to the Browse tab within Segments and toggle on Add all segments to schedule. This will cause all segments to be evaluated based on the schedule set by your organization.
Scheduled evaluation can be enabled for sandboxes with a maximum of five (5) merge policies for XDM Individual Profile. If your organization has more than five merge policies for XDM Individual Profile within a single sandbox environment, you will not be able to use scheduled evaluation.
Schedules can currently only be created using the API. For detailed steps on creating, editing, and working with schedules using the API, please follow the tutorial for evaluating and accessing segment results, specifically the section on scheduled evaluation using the API.
The audiences functionality is currently in limited beta and is not available to all users. The documentation and the functionality are subject to change.
Select the Audiences tab to see a list of all the audiences for your organization.
By default, this view lists information about the audiences including the name, profile count, origin, created date, and last modified date.
You can select the icon to change which fields are displayed.
A popover appears, listing all the fields that can be displayed within the table.
Field | Description |
---|---|
Name | The name of the audience. |
Profile count | The total number of profiles that qualify for the audience. |
Origin | The origin of the audience. If this audience was Platform-generated, it will have an origin of Segmentation Service. |
Lifecycle status | The status of the audience. Possible values for this field include Draft , Published , and Archived . |
Update frequency | A value that states how often the audience’s data is updated. Possible values for this field include On Demand , Scheduled , and Continuous . |
Last updated by | The name of the person who last updated the audience. |
Created | The time and date the audience was created. |
Last updated | The time and date the audience was last created. |
Access labels | The access labels for the audience. Access labels allow you to categorize datasets and fields according to usage policies that apply to that data. These labels can be applied at any time, providing flexibility in how you choose to govern data. For more information on access labels, please read the documentation on managing labels. |
You can select Create Audience to create an audience.
A popover appears, letting you choose between composing an audience or building rules.
Selecting Compose Audiences takes you to the Audience Builder. To learn more about creating audiences, please read the Audience Builder guide.
Selecting Build Rule takes you to the Segment Builder. To learn more about creating segments, please read the Segment Builder guide
To see more details about a specific audience, select an audience’s name within the Audiences tab.
The audience details page appears. This page differs in details depending on whether the audience was generated with Adobe Experience Platform or from an external source such as Audience Orchestration.
For more information on Platform-generated audiences, please read the segment summary section.
On the top of the audience details page, there is a summary of the audience and details about the dataset the audience is saved in.
The Audience summary section provides information such as the ID, name, description, and details of the attributes.
The Dataset details section provides information such as the name, description, table name, source, and schema. You can select View dataset to see more information about the dataset.
Field | Description |
---|---|
Name | The name of the dataset. |
Description | The description of the dataset. |
Table name | The table name of the dataset. |
Source | The source of the dataset. For externally generated audiences, this value will be Schema. |
Schema | The type of XDM schema that the dataset corresponds to. |
To learn more about datasets, please read the dataset overview.
Streaming segmentation is the ability to do segmentation on Platform in near real-time, while focusing on data richness. With streaming segmentation, segment qualification now happens as data lands into Platform, alleviating the need to schedule and run segmentation jobs.
More information about streaming segmentation can be found in the streaming segmentation user guide.
In order for streaming segmentation to work, you will need to enable scheduled segmentation for the organization. For details on enabling scheduled segmentation, please refer to the streaming segmentation section in this user guide.
Edge segmentation is the ability to evaluate segments in Platform instantaneously on the edge, enabling same page and next page personalization use cases.
More information about edge segmentation can be found in the edge segmentation UI guide
Policy violations only apply if you are creating a segment that has been assigned to a destination.
Once you are done creating your segment, the segment will be analyzed by Adobe Experience Platform Data Governance to ensure there are no policy violations within the segment. See the Data Governance overview for more information.
The Segmentation Service UI provides a rich workflow allowing you to isolate marketable audiences from Real-Time Customer Profile data.
To learn more about Segmentation Service, please continue reading the documentation. To learn how to use the Segmentation Service API, please read the Segmentation Service developer guide.