Excluding Traits in Algorithmic (Look-Alike) Models
In this video you will learn how and why to exclude specific (or groups of) traits from an Algorithmic (Look-Alike) Model.
Use cases for this feature include:
- Extremely common traits such as site visitor traits bias the model which won’t be useful in finding a quality look-alike audience. Customers no longer have to create a separate data source and store common traits in the new data source, but can now simply exclude them.
- There is now a way to use a subset of traits from a third party, such as just behavioral interests, rather than all the information in a model. Third parties usually send a lot of data which might not be useful for the customer. In some cases, they won’t be allowed to use all of the data from a legal perspective in modeling. Now you can exclude traits or folders of traits that you don’t want to include in the model.