Implementation guide for identity graph linking rules

Read this document for a step-by-step by guide that you can follow when implementing your data with Adobe Experience Platform Identity Service.

Step-by-step outline:

Prerequisites for implementation prerequisites-for-implementation

This section outlines prerequisite steps that you must complete prior to implementing identity graph linking rules to your data.

Unique namespace

Single person namespace requirement single-person-namespace-requirement

You must ensure that the unique namespace with the highest priority is always present in every profile. Doing so allows Identity Service to detect the appropriate person identifier in a given graph.

Namespace priority configuration

If you are using the Adobe Analytics source connector to ingest data, then you must give your ECIDs a higher priority than Adobe Analytics ID (AAID) because Identity Service blocks AAID. By prioritizing ECID, you can instruct Real-Time Customer Profile to store unauthenticated events to ECID instead of AAID.

XDM experience events

During your pre-implementation process, you must ensure that the authenticated events that your system will send to Experience Platform always contain a person identifier, such as CRMID.

{
  "_id": "test_id",
  "identityMap": {
      "ECID": [
          {
              "id": "62486695051193343923965772747993477018",
              "primary": false
          }
      ],
      "CRMID": [
          {
              "id": "John",
              "primary": true
          }
      ]
  },
  "timestamp": "2024-09-24T15:02:32+00:00",
  "web": {
      "webPageDetails": {
          "URL": "https://business.adobe.com/",
          "name": "Adobe Business"
      }
  }
}

{
    "_id": "test_id",
    "identityMap": {
        "ECID": [
            {
                "id": "62486695051193343923965772747993477018",
                "primary": false
            }
        ]
    },
    "timestamp": "2024-09-24T15:02:32+00:00",
    "web": {
        "webPageDetails": {
            "URL": "https://business.adobe.com/",
            "name": "Adobe Business"
        }
    }
}

Do not send an empty string as an identity value when sending events using XDM experience events. If the identity value of the namespace with highest namespace priority is an empty string, the record will be ignored from Real-Time Customer Profile. This applies to both identityMap, as well as fields marked as an identity.

    "identityMap": {
        "ECID": [
            {
                "id": "24165048599243194405404369473457348936",
                "primary": false
            }
        ],
        "Phone": [
            {
                "id": "",
                "primary": true
            }
        ]
    }

You must ensure that you have a fully qualified identity when sending events using XDM experience events.

    "identityMap": {
        "ECID": [
            {
                "id": "24165048599243194405404369473457348936",
                "primary": false
            }
        ]
    }

Set permissions set-permissions

The first step in the implementation process for Identity Service is to ensure that your Experience Platform account is added to a role that is provisioned with the necessary permissions. Your administrator can configure permissions for your account by navigating to the Permissions UI in Adobe Experience Cloud. From there, your account must be added to a role with the following permissions:

For more information on permissions, read the permissions guide.

Create your identity namespaces namespace

If your data requires it, you must first create the appropriate namespaces for your organization. For steps on how to create a custom namespace, read the guide on creating a custom namespace in the UI.

Use graph simulation tool graph-simulation

Next, navigate to the graph simulation tool in the Identity Service UI workspace. You can use the graph simulation tool to simulate identity graphs, built with a variety of different unique namespace and namespace priority configurations.

By creating different configurations, you can use the graph simulation tool to learn and better understand how the identity optimization algorithm and certain configurations can affect how your graph behaves.

Configure identity settings identity-settings

Once you have a better idea of how you want your graph to behave, navigate to the identity settings UI in the Identity Service UI workspace. To access the identity settings UI, select Identities from the left navigation and then select Settings.

Use the identity settings UI to designate your unique namespaces and configure your namespaces by order of priority. Once you are finished with applying your settings, you must wait at least six hours before you can proceed to ingest data, as it takes at least six hours for new settings to be reflected in Identity Service.

For more information, read the identity settings UI guide.

Create an XDM schema schema

With your unique namespaces and namespace priorities established, you can now proceed to required set up in order to ingest your data. First, you must create an XDM schema. Depending on your data, you may need to create a schema for both XDM Individual Profile and XDM ExperienceEvent.

To ingest data into Real-Time Customer Profile, you must ensure that your schema contains at least one field that has been designated as the primary identity. By setting a primary identity, you can enable a given schema for Profile ingestion.

For instructions on how to create a schema, read the guide on creating an XDM schema in the UI.

Create a dataset dataset

Next, create a dataset to provide a structure for the data that you are going to ingest. A dataset is a storage and management construct for a collection of data, typically a table, that contains a schema (columns) and fields (rows). Datasets work in tandem with schemas, and to ingest data to Real-Time Customer Profile, your dataset must be enabled for Profile ingestion. In order for your dataset to be enabled for Profile, it must reference a schema that is enabled for Profile ingestion.

For instructions on how to create a dataset, read the dataset UI guide.

Ingest your data ingest

By this point, you should have the following:

Once you have all of the items listed above, then you can begin ingesting your data to Experience Platform. You can perform data ingestion through several different ways. You can use the following services to bring your data to Experience Platform:

For any feedback, use the Beta feedback option in the Identity Service UI workspace.

Validate your graphs validate

Use the identity dashboard for insights on the state of your identity graphs, such as your overall identity count and graph count trends, identity count by namespace, and graph count by graph size. You can also use the identity dashboard to view trends on graphs with two or more identities, organized by namespace.

Select the ellipses (...) and then select View more for further information and to validate that there are no collapsed graphs.

Use the window that appears to view information on your collapsed graphs. In this example, both email and phone are marked as unique namespace, so therefore, there are no collapsed graphs in your sandbox.

Appendix appendix

Read this section for additional information that you can refer to when implementing your identity settings and unique namespaces.

Dangling loginID scenario dangling-loginid-scenario

The following graph simulates a “dangling” loginID scenario. In this example, two different loginIDs are bound to the same ECID. However, {loginID: ID_C} is not linked to the CRMID. Therefore, there is no way for Identity Service to detect that these two loginIDs represent two different entities.

Next steps

For more information on identity graph linking rules, read the following documentation: