In this article
Using Vaults to add a new Data Source.
Data Sources is where you can set up and upload a new data import to create Matched or Modeled cohorts from that data source.
Add a New Data Source
In the Audience Platform Navigation Bar:
1. Look for Vaults
2. Go to the Data Sources Tab
3. Click '+ Add Source' for the builder to appear
4. Name the Data Source
5. Give any description that fits
6. Under GCS account email, type in the email address associated with the Google Storage Cloud account you want to use to upload the data.
7. Next, you will want to choose the taxonomy to upload. The taxonomy file is a CSV file that contains a mapping of the Audience ID to the audience name. The file consists of two columns:
- The first column should list the ID. This ID is what is received as part of the user data upload.
- The second column should list the name. This name is what will be used to display the Cohort name in the dashboard.
8. Once the file has successfully uploaded you will click 'Create' and a message will pop up letting you know the data source was created and share the location of the GCS bucket where we’ll import the data in.
If you navigate back to the 'Data Sources' tab, you will be able to see the recently created sources and check all the relevant information associated.
Example: The ID, Name, and GCS bucket details.
Send User Data
Once the data source has been configured you can start to send that user data into Permutive via the GCS file. The data needs to be sent to Permutive as a TCV file with three separate columns:
1. The first column should list the user ID
2. The second column should list the ID type which can be a hashed email, IDFA, Appnexus ID, or a third-party cookie. We also support all different types of ID’s in one singular file.
3. The third column should list the audiences. This list is in comma separated format of individual cohort ID’s that the user belongs to. This will reflect the ID’s you have uploaded in the taxonomy file.
4. Once the advertiser drops that file into the GCS bucket, Permutive will start the import process which transforms this user data file into a data sketch and then deletes the original file.
Tip: You can see an overview of your data sources which includes the status of the import, how many records have been imported, and the date of the last import.