Connector
Pinterest Ads data, in the full picture.
Pinterest drives discovery and saves that convert later, which is exactly when platform attribution undercounts. Odins syncs campaign analytics daily and structures Pinterest next to the rest of your mix, so the model can measure its real contribution.
Fully managed · reviewed September 2026
The data
What you can get from Pinterest.
Everything the platform's API reports is available. The default set below covers daily performance; the rest is added on request.
Stream names follow the managed pipeline layer's Pinterest source. Full reference: Airbyte's pinterest documentation and Pinterest's API docs.
Discovery spend, measured properly.
A save today becomes a purchase next month. Modeling against sales data catches that; last-click never will.
- OAuth connection, multiple ad accounts in one connector
- Ad group structure preserved for your own analysis
- Daily incremental syncs with deduplication, monitored by our team
- Raw tables and the structured dataset both deliverable to your warehouse
Frequently asked questions.
How do I get Pinterest Ads data into our data warehouse?
Connect Pinterest to Odins with OAuth and we sync campaign analytics daily into a managed pipeline, then onward to BigQuery, Snowflake, Redshift or Databricks.
Pinterest converts late. How do you measure that?
That is what the marketing mix model is for. It estimates Pinterest's contribution against sales data over time, including delayed effects, instead of relying on a click window that closes before the purchase happens.
Will the numbers match Pinterest Ads Manager?
Yes, within Pinterest's own reporting variance. We pull from the same API, refreshed daily.
How detailed is the data?
Daily per campaign by default, with ad group and ad account structure preserved. More detailed API reports can be added to your sync.
How much history do you load?
2 to 3 years where the API allows it, loaded during onboarding.
Marketing spend, modeled like an investment.
A 30-minute walkthrough of the connection, the data and what the model does with it.
Book a demo-1.png?width=1132&height=292&name=Asset%201@2x%20(12)-1.png)