Skip to main content

[Private Preview] Matching on Client-Encoded RampIDs in Clean Rooms

LiveRamp clean rooms now allow for seamless translation between client-encoded RampIDs within hybrid clean rooms without needing to alter the code in the questions. Historically, leveraging RampIDs required joins to mapping files between client-encoded RampIDs in question code in order to traverse between RampID universes. With RampID matching in hybrid clean rooms, clients can now assign RampID-keyed datasets to questions, join on RampID in the SQL code, and LiveRamp will take care of matching between parties, with no mapping files necessary.

Note

  • This feature is currently in private preview. Contact your customer success representative to be included in the program.

  • This feature is only available for hybrid and confidential computing (TEE) clean rooms.

Overall Steps

Perform the following overall steps to enable matching on RampID in clean rooms:

  1. Configure the data connection and indicate which fields are RampID fields.

  2. Configure RampID-keyed datasets in applicable clean rooms.

  3. Assign datasets to questions using RampID-keyed datasets.

For information on performing these steps, see the sections below.

For information on the limitations of the Private Preview, see the "Limitations of the Private Preview" section below.

Prerequisites

  • Datasets must already be matched to RampID on your own encoding before they are connected to your organization as data connections. The recommended data structure is:

    • md5_hashed(CID) | Ramp_ID

    • md5_hashed(CID) | attribute_1 | attribute_2 | etc.

    Note

    Contact your LiveRamp representative for guidance if you do not have a CID, the ability to hash your CID, or if you need assistance considering the optimal data structure for RampID matching in clean rooms.

  • A clean room can contain either datasets keyed off PII (such as hashed email) or RampIDs. This is required to maintain the pseudonymity of the RampIDs.

Configure the Data Connection

For each data connection to be configured, follow the instructions for the appropriate data connection type in "Cloud-Based Data Connections", with these additional steps:

  1. For the "Map the Fields" step, slide the User Identifiers toggle to the right for the column containing RampIDs, and then select RampID from the list.

  2. You will be prompted to enter a name for your RampID encoding and provide the RampID domain identifier used to encode the RampIDs for the associated data connection. This is a 4-digit integer which can be inferred from the prefix used in the RampIDs (i.e. XY1234 --> domain ID = 1234).

    Note

    • If you are unsure what your domain ID is, contact your LiveRamp representative.

    • An organization can configure one domain ID per organization. If your team leverages more than one RampID encoding, contact your LiveRamp representative.

  3. Once you've configured your domain ID, save the mapping. You can always refer to or update your domain ID in Clean Room's Data Management → Identity section.

Configure the Datasets

To use RampIDs as match keys in a clean room, configure the datasets:

Note

The RampID domain used for your organization in the clean room can be referenced on the Details page.

  1. From the clean room, select Datasets from the navigation pane.

  2. Configure your RampID-keyed dataset(s).

  3. LiveRamp will detect your inclusion of a RampID field and ask you to affirmatively acknowledge you are comfortable using RampID as a join key in the clean room and abide by LiveRamp's requirements to maintain pseudonymity of the RampID. Once confirmed, the use of the domain associated with this RampID will be locked for the clean room collaboration.

You are now ready to match on RampIDs in clean room questions.

Assign Datasets to Questions

Once questions have been provisioned to or authored within the clean room, you can assign datasets to them to match on RampIDs:

  1. From the Questions page, click Manage Datasets for the desired question.

  2. Assign your RampID-keyed dataset to the question.

  3. As required, map field macros to the macros associated with the question. This may include mapping your RampID field to the macro expected to be used for matching in the question.

  4. Save your assignment.

  5. When ready, select Report → New Report for the question. LiveRamp will detect RampIDs are in use and perform the translation between RampID encodings on your behalf.

Note

The matching step slows performance for question runs, but there should not be a significant increase in query run time. Most of the runtime will be spent on actual calculations, and matching on RampID reduces the overhead previously required when matching on a mapping file.

Limitations of the Private Preview

  • List questions will output the RampID associated with the clean room owner's organization. Do not configure list questions if you want a partner's RampID to be used for export or activation.

    Note

    Only one RampID will be written to the output of a list question. Do not attempt to construct mapping files between RampIDs in a List question.

  • LiveRamp does not yet facilitate matching between RampIDs for clean compute questions.

  • Work is in progress to enable seamless matching on RampIDs for cloud-native clean room types.

In this section: