Data-driven attribution models

If you’re new to attribution, we recommend using data-driven models. Since they're based on consumer behavior, you'll benefit from attribution reporting without studying the default algorithms or investing in user research. Also, natural search traffic is eligible to receive credit in data-driven models, which can give a more complete picture of conversion data.

Due to differences in cookies and device ID settings, there may be times when conversions can't be measured directly. In some cases, Campaign Manager 360 uses machine learning and historical data to model conversions.

Create a new data-driven attribution model

1. Go to Reporting & Attribution > Attribution.
2. Choose a Floodlight configuration. All impressions and clicks tracked through this configuration will be taken into consideration by the model.
3. Click Attribution Modeling Tool.
4. A default model appears. Click dropdown next to the default model, at the bottom of the list click Create new data-driven model.
5. Name your model. Tip: Include today's date.
6. To train the data-driven model correctly, select all Floodlight conversions that represent key customer actions such as purchases. You can filter to a single Floodlight conversion later.
   • Note: Don't include Floodlight conversions that were created solely for building audience lists such as remarketing pixels.
7. Choose Basic Channel Grouping (for most setups).
   1. To create a custom channel grouping go to: Reporting & Attribution > Attribution > Attribution Modeling Tool > Channel Groupings.
8. (Optional) Set a custom lookback window.
9. Click Save.

You can only create one data-driven attribution model. Your model needs to train for 1 to 2 days, and will be greyed out until usable. If you make changes to an existing model, the model will need to retrain, which may take up to 9 days. Reporting is unavailable for date ranges prior to a model being trained.