Export data from Campaign to Adobe Experience Platform sources

To export Campaign Classic data to Adobe Real-time Customer Data Platform (RTCDP), you first need to build a workflow in Campaign Classic to export to your S3 or Azure blob storage location the data that you want to share.

Once the workflow has been configured and data sent to your storage location, you need to connect your S3 or Azure blob storage location as a Source in Adobe experience Platform.

NOTE
Please note that we recommend exporting Campaign generated data only (e.g. sends, opens, clicks etc.) to Adobe Experience Platform. Data that is ingested from a 3rd party source (like your CRM) should be imported directly into Adobe Experience Platform.

Create an export workflow in Campaign Classic

To export data from Campaign Classic to your S3 or Azure Blob storage location, you need to build a workflow to target the data you want to export and send it to your storage location.

To do this, add and configure:

  • An Data extraction (file) activity to extract the targeted data into a CSV file. For more on how to configure this activity, refer to this section.

  • A File transfer activity to transfer the CSV file to your storage location. For more on how to configure this activity, refer to this section.

As an example, the workflow below extracts logs on a regular basis into a CSV file, then transfers the file to a storage location.

Connect your storage location as a Source

The main steps to connect your S3 or Azure blob storage location as a Source in Adobe experience Platform are listed below. Detailed information on each of these steps is available in the Source connectors documentation.

  1. In Adobe Experience platform Sources menu, create a connection to your storage location:

    note note
    NOTE
    The storage location can be Amazon S3, SFTP with Password, SFTP with SSH Key, or Azure Blob connections. The preferred method to send data to Adobe Campaign is through Amazon S3 or Azure Blob:

  2. Configure a dataflow for a cloud storage batch connection. A dataflow is a scheduled task that retrieves and ingests data from the storage location to an Adobe Experience Platform dataset. This steps allows you to configure the data ingestion from your storage location, including data selection and the mapping of the CSV fields to an XDM schema.

    Detailed information is available in this page.

  3. After the Source has been configured, Adobe Experience Platform will import the file from the storage location that you provided.

    This operation can be scheduled according to your needs. We recommend performing the export up to 6 times a day, depending on the load already present on the instance.

recommendation-more-help
601d79c3-e613-4db3-889a-ae959cd9e3e1