Look-Alike Modeling: Trait Exclusion
Trait Exclusion provides additional controls in your modeling workflow, allowing you to add the necessary guard rails to the model, based on your domain expertise and regulatory requirements. Use the Exclusions option to select which traits to ignore when creating models from one or more data sources.
Use Cases
Here are some use cases you can address with Trait Exclusion:
- Trait Exclusion enables you to exclude certain catch-all traits, such as site visitor traits, so you don’t bias the model, leading to flat results.
- You can remove traits that you don’t know about or you don’t trust from a data source, to better understand the influential traits.
- You can exclude certain traits, such as demographic traits, to help with any compliance obligations you may have.
How to Use Trait Exclusions
Use the Build a model workflow to build new algorithmic models.
- The Exclusions selection is greyed out until you select one or more data sources for modeling.
- After selecting one or more data sources for modeling, press Browse All Traits.
- In the Select Traits to Exclude window, you can see all traits associated with the data sources you selected previously. Select the traits you want to exclude.
- You can filter the traits by trait type, trait population type (Device ID and Cross-Device ID), or you can browse the trait folders. Note that trait folders only display the traits associated with your selected data sources.
- Press Exclude Selected Traits.
If you prefer video tutorials, watch our video demonstration for Trait Exclusion:
Additionally, watch the video below for a detailed look at how cross-device metrics work.
Important Aspects & Limitations
Please take note of the following aspects and limitations related to Trait Exclusion:
Note the following limitations for companies using RBAC:
- If you don't have access to view a trait, you cannot select that trait to be excluded from the model.
- If you don't have access to view a trait, you cannot view that trait in the excluded traits list.
Watch the video below to learn how and why to exclude specific traits from a Look-Alike Model.