DMP Segment Uploader

This document is for Data Management Platform (DMP) segment providers who wish to include their DMP segments in TAP's Audience Segments Targeting feature. It describes the process and specifications used for uploading DMP targeting segments to Triton Digital.

Overview

• Publishers who use your DMP segments must have a contract with Triton Digital that provides them access to your DMP audience targeting segments.

• You must work with a Triton Digital Partner Integration Specialist to initiate the upload process and to confirm the setup.

• The S3 folders described below are set up by Triton Digital, who will provide you with access details once they are set up.

• Initially, your segment uploads will be evaluated by a Triton Digital Partner Integration Specialist to confirm they conform to specifications. If they do not conform, the Triton Digital Partner Integration Specialist will work with you to help fix any errors. Once everything is “on spec”, your uploads can be done automatically.

Taxonomy

Taxonomy File Locations

• Two S3 buckets are initiated by Triton Digital; one for production and one for pre-production:

  • s3://triton-dmp-integrations>{environment(prod|preprod)}>{DMP Provider}>{PublisherID}>taxonomy>

• Files should be dropped in a folder representing the day of the file under the format (YYYYMMDD).

• Files should be placed in a specific folder for each day:

  • s3://triton-dmp-integrations>{environment(prod|preprod)}>{DMP Provider}>{PublisherID}>taxonomy>{YYYYMMDD}

Example for DMP provider DoccoSegments, publisher ID 12345, on "prod" environment:

s3://triton-dmp-integrations>{environment(prod)}>{DoccoSegments}>{12345}>taxonomy>{20230206}

Taxonomy File Format

The naming convention for a taxonomy file is as follows:

YYYYMMDD.VOLUME_NUMBER.taxonomy.tsv.gz

• YYYYMMDD: Date of upload.

• VOLUME_NUMBER: Volume number to split a large file into many smaller ones

• Example: 20220803.001.taxonomy.tsv.gz

The taxonomy file is a tab-separated formatted file (tsv) which includes the following columns in the following order:

• Segment ID

  • Required

  • Must be unique for the publisher being integrated

• Segment Name

  • Required

• Price

  • Required.

  • Must be in USD. Decimals are optional. Example: 2 is interpreted as 2.00, and 2.00 is also interpreted as 2.00.

  • 0 is an accepted value

  • Double (data type)

• Company

  • Optional

  • Can be used if the segment is provided by a third party

• Segment Category

  • Optional

• Status

  • Required

  • Active or Inactive

Example:

Segment ID        Segment Name        Price        Company        Segment Category  Status
111   Age > 25-30   1.40   Triton Digital   Age Active

Taxonomy Processing Details

• Each segment is updated completely with provided data.

• If an existing segment is not in the file, it will not be modified.

Segments for Listener Mapping

Segment File Locations

• Two S3 buckets are initiated by Triton Digital; one for production and one for pre-production:

  • s3://triton-dmp-integrations>{environment(prod|preprod)}>{DMP Provider}>{PublisherID}>segments>

• Files should be dropped in a folder representing the day of the file under the format (YYYYMMDD).

Full Segment File Format

Use a full segment file to completely replace the listener database with the contents of the new segments file.

The naming convention for a full segment file is as follows:

full.YYYYMMDD.VOLUME_NUMBER.TYPE.tsv.gz

• YYYYMMDD: Date of upload

• VOLUME_NUMBER: Volume number to split a large file into many smaller ones

• TYPE: Listener ID type, which is one of cookie, gaid, idfa, or ip

• Example: full.20220803.001.cookie.tsv.gz

A full segment file is a tab-separated formatted file (tsv) with the following columns in the following order:

• listener-id

• segment-ids: Comma-separated list of segment IDs

Example:

listener-id   segment-ids
90468516-427e-453b-b8f7-fee93c539f76   111,222,333,444

Incremental Segment File Format

Use an incremental segment file to add or remove the defined segments in the new segments file.

The naming convention for an incremental segment file is as follows:

inc.YYYYMMDD.VOLUME_NUMBER.TYPE.tsv.gz

• YYYYMMDD: Date of upload

• VOLUME_NUMBER: Volume number to split a large file into many smaller ones

• TYPE: Listener ID type, which is one of cookie, gaid, idfa, or ip

• Example: inc.20220803.001.cookie.tsv.gz

An incremental segment file is a tab-separated formatted file (tsv) with the following columns in the following order:

• listener-id

• segment-ids: Comma-separated list of segment IDs

• Start a line with + to add the defined segments.

• Start a line with - to remove the defined segments.

Example:

listener-id   segment-ids
+90468516-427e-453b-b8f7-fee93c539f76   222,333
+6158409-e724-b35b-7f8b-76f935c39eef   111

Segment Processing Details

• Each time a file is processed, the segments previously imported for any given listener are overwritten. The segments of listeners not present in the file being processed are not changed.

• If no changes are made to a listener within 35 days, their entry in the database is erased. Therefore if you (the DMP) want to keep the segments active for a listener, you need to send it again within 35 days.

• Segments not uploaded via the taxonomy uploader are ignored.