This article describes the setup steps for Criteo → Google BigQuery pipeline.
1. On the OWOX BI main page, click the 'New' button and choose 'Pipeline':
2. As a service provider, select 'Criteo':
3. Choose pipeline 'Criteo → Google BigQuery':
And click the 'Create & Setup' button:
4. Grant OWOX BI access to your Criteo portfolio. To do it, click Add. Then on the Access request page, select only one portfolio and click Approve.
To grant access, you should be assigned the Administrator role in the required portfolio.
NoteYou should grant access to only one Criteo portfolio per OWOX BI pipeline. To integrate OWOX BI with several Criteo portfolios, follow these instructions.
5. Select or add new access to a Google BigQuery account where you want to store the exported data:
6. Select a Google BigQuery project and create a dataset you want to upload your data to (or choose an existing one):
NoteTo set up the data collection, your Google account must be granted both the BigQuery Data Editor and BigQuery User roles for the destination project. Otherwise, BigQuery won’t let you upload the data.
To check/grant the permissions, go to the Identity and Access Management page in your Google Cloud Platform project. Read more in Google documentation.
7. Specify the settings for your pipeline:
- The date that marks the beginning of the period you want to upload cost data for. You can set up either the future or past date. In the latter case, learn the existing limitations on historical data import first.
- The currency you want your cost data to be additionally converted into. The converted costs are useful for the reports where you need to have cost data from different sources presented in a uniform currency. The pipeline uploads ad cost data into two separate fields in your BigQuery data table. One field contains costs in the currency of data import (USD). Another field contains costs converted into the currency you select when creating the pipeline. OWOX BI uses API Open Exchange Rates for currency conversion.
- Values for source/medium you want to apply by default to the imported data.
Make sure the source/medium values you provided here correspond to UTM tags in your ad account. Otherwise, data discrepancy occurs during the session cost calculation. Check this article for the details.
At any time, you can change the source/medium settings on the pipeline page.
Note that the changes will be applied to the newly imported data. Historical data will be updated with the new source/medium values within the preset actualization window only.
- Also, you can connect your new pipeline with the Cost Data pipeline to attribute the imported ad cost to user sessions. Read more
8. Click Create pipeline.
Done! The pipeline will be uploading data to a partitioned table in the selected BigQuery dataset. The data will be uploaded daily — for the previous day.
OWOX BI updates the imported data during the preset actualization window if cost data changes retrospectively in your ad account.
The table data schema is in this article.