Ingest and use Experience Platform audiences

In the Experience Platform UI:

1. Select Customer > Audiences.

2. Select Browse and search for the audiences that you want to ingest and use in Customer Journey Analytics. Note the Audience Id for each of the audiences for later use.

   

Request

curl -X POST \
  'https://platform.adobe.io/data/foundation/catalog/dataSets?requestDataSource=true' \
  -H 'Authorization: Bearer {ACCESS_TOKEN}' \
  -H 'Content-Type: application/json' \
  -H 'x-api-key: {API_KEY}' \
  -H 'x-gw-ims-org-id: {ORG_ID}' \
  -H 'x-sandbox-name: {SANDBOX_NAME}' \
  -d '{
   "name": "{DATASET_NAME}",
   "schemaRef": {
      "id": "_xdm.context.profile__union",
      "contentType": "application/vnd.adobe.xed+json;version=1"
   },
   "fileDescription": {
      "persistet": true,
      "containerFormat": "parquet",
      "format": "parquet"
   }
}'

Where:

• DATASET_NAME is the friendly name of the dataset. For example, Segment Export Job Dataset for CJA.

Response

["@/dataSets/{DATASET_ID}"]

Where:

• DATASET_ID is the dataset identifier for the created dataset.

curl -X POST https://platform.adobe.io/data/core/ups/export/jobs \
 -H 'Authorization: Bearer {ACCESS_TOKEN}' \
 -H 'Content-Type: application/json' \
 -H 'x-gw-ims-org-id: {ORG_ID}' \
 -H 'x-api-key: {API_KEY}' \
 -H 'x-sandbox-name: {SANDBOX_NAME}' \
 -d '{
    "fields": "{COMMA_SEPARATED_LIST_OF_FULLY_QUALIFIED_FIELD_NAMES}",
    "filter": {
        "segments": [
            {
                "segmentId": "{AUDIENCE_ID_1}",
                "segmentNs": "ups",
                "status": [
                    "realized"
                ],
                "segmentId": "{AUDIENCE_ID_2}",
                "segmentNs": "ups",
                "status": [
                    "realized"
                ],
                "segmentId": "{AUDIENCE_ID_3}",
                "segmentNs": "ups",
                "status": [
                    "realized"
                ]
             }
        ]
    },
    "destination":{
        "datasetId": "{DATASET_ID}",
        "segmentPerBatch": false
    },
    "schema":{
        "name": "_xdm.context.profile"
    }
}'

Where

• COMMA_SEPARATED_LIST_OF_FULLY_QUALIFIED_FIELD_NAMES could be something like _demoemea.identification.core.ecid, _demoemea.identification.core.email, _demoemea.identification.core.phoneNumber, person.gender, person.name.firstName, person.name.lastName. Ensure you include at least the relevant fields (like the personID (email)) that you want to use in your Customer Journey Analysis.
• AUDIENCE_ID_x are the audience identifiers of the audiences that you want to export.
• DATASET_ID is the dataset that you created.

Response

{
  "..."
  "id": "{EXPORT_JOB_ID}",
  "..."
}

Where

• EXPORT_JOB_ID is the identifier of the export job.

Request

curl -X GET https://platform.adobe.io/data/core/ups/export/jobs/{EXPORT_JOB_ID} \
 -H 'Authorization: Bearer {ACCESS_TOKEN}' \
 -H 'x-gw-ims-org-id: {ORG_ID}' \
 -H 'x-api-key: {API_KEY}' \
 -H 'x-sandbox-name: {SANDBOX_NAME}'

Response

{
  "..."
  "id": "{EXPORT_JOB_ID}",
  "..."
  "status": "SUCCEEDED",
  "..."
}

In the Experience Platform UI:

1. Select Data Management > Datasets.

2. Select the dataset that you created, for example: Segment Export Job Dataset for CJA.

   

3. Verify the ingested batches. If the dataset contains failed batches, use Data Management > Monitoring to see what is the reason. For example, you used a field name that does not exist in the schema.

4. Copy the Table name of the dataset. For example: segment_export_job_dataset_for_cja. You use that name in the next step.

Use a PSQL client that connects to Experience Platform Query Service.

In the Experience Platform UI:

1. Select Data Management > Queries.
2. Select Credentials.

Use the credentials to configure your PSQL client to connect to Customer Journey Analytics Query Service.

Query

Execute this query to retrieve the audience data from the dataset:

SELECT ROW_NUMBER() OVER (ORDER BY key)::text as _id, personID, key as audienceMembershipId
FROM (
   SELECT {IDENTITY_TO_USE_AS_PERSON_ID} AS personID, explode(segmentMembership.ups)
   FROM {DATASET_TABLE_NAME}
)
WHERE value.status = 'realized' AND (key = '{AUDIENCE_ID_1}' OR key = 'AUDIENCE_ID_2' OR key = 'AUDIENCE_ID_3')

Where:

• IDENTITY_TO_USE_AS_PERSON_ID is one of the fields you defined as part of the export job. For example: _demoemea.identification.core.email.
• DATASET_TABLE_NAME is the table name of the dataset.
• AUDIENCE_ID_x are the audiences you defined as part of the export job. You need to specify these audiences once more as the specification in the export job is a row-level filter. That row-level filter returns profiles for the specified segments with all the segment memberships for each of the profiles.

Results

The result of the query, in JSON format, should look like:

[
   {
      "_id": "1",
      "personID": "{PERSON_ID_x}",
      "audienceMembershipId": "{AUDIENCE_ID_x}"
   },
   {
      "_id": "2",
      "personID": "PERSON_ID_y",
      "audienceMembershipId": "{AUDIENCE_ID_x}"
   }

]

Where:

• PERSON_ID_x are the identifier values for the identifier you want to use as the person ID. For example, john.doe@gmail.com when you use email.
• AUDIENCE_ID_x are the audience identifiers.

The final JSON should look like:

[
   {
      "_id": "1",
      "personID": "{PERSON_ID_x}",
      "{TENANT_NAME}": {
         "audienceMembershipId": "{AUDIENCE_ID_x}",
         "audienceMembershipName": "{AUDIENCE_FRIENDLY_NAME_x}"
      }
  },
  {
      "_id": "2",
      "personID": "{PERSON_ID_y}",
      "{TENANT_NAME}": {
         "audienceMembershipId": "{AUDIENCE_ID_y}",
         "audienceMembershipName": "{AUDIENCE_FRIENDLY_NAME_y}"
      }
    }
  }

]

Where:

• TENANT_NAME is the name of the tenant. For example: _demoemea.
• PERSON_ID_x are the identifier values for the identifier that you want to use as the person ID. For example, john.doe@gmail.com when you use email.
• AUDIENCE_ID_x are the audience identifiers.
• AUDIENCE_FRIENDLY_NAME_x are friendly audience names for the audience ids. For example: Luma - Blue+ Members.

Use your favorite tool to transform the original JSON to this format.

To create the schema:

In the Experience Platform UI:

1. Select Data Management > Schemas.

2. Select Create schema. Select Standard from the drop-down menu.

3. Select Manual in the Create a schema dialog and use Select to continue.

4. In the Create schema wizard, in the Select a class step:

   1. Select Individual Profile.
   2. Select Next.

5. In the Create schema wizard, in the Name and review step:

   1. Enter a Schema display name. For example: Audience Export for CJA Schema.
   2. (optional) Enter a Description.
   3. Select Finish.

6. Set up your schema to contain a custom field group (named, for example, Audience Membership) that contains two fields named audienceMembershipId and audienceMembershipName.

7. Ensure that the personID field is an Identity, Primary Identity and has Email as the Identity namespace.

   

8. Apply all changes. Select Save to save the schema.

In the Experience Platform UI:

1. Select Data Management > Datasets.

2. Select Create dataset.

3. Select Create dataset from schema.

4. In the Create dataset from schema wizard, in the Select schema step:

   1. Select the schema you just created. For example: Audience Export for CJA Schema.
   2. Select Next.

5. In the Create dataset from schema wizard, in the Configure dataset step:

   1. Enter a Name for the dataset.
   2. (optional) Enter a Description for the dataset.
   3. Select Finish.

6. In the Datasets > name of the dataset, drag the transformed JSON data file and drop the file onto Drag and drop files. This action starts the ingestion of the exported JSON data into the dataset.

7. Verify the ingested batches. If the dataset contains failed batches, use Data Management > Monitoring to see what is the reason. For example, you defined a field name in the JSON that does not exist in the schema.

In the Customer Journey Analytics UI:

1. Select Data Management > Connections.

2. Create a new connection/ Define Connection settings and Data settings. Or select an existing connection and use Edit Connection to edit the connection.

3. Select Add datasets.

4. Select the dataset that you created and in which you ingested the transformed JSON data.

5. Configure the dataset. For example:

   

6. Save the connection.

1. Select Data Management > Data views.

2. Edit an existing data view or create a new data view.

3. In the Components tab of the data view, ensure Audience Membership Id and Audience Membership Name are added as dimension components.

   

4. Select Save and Continue to save the data view.