Recommenders
Recommender predictions allow you to predict which product, service, or other offer that any individual is likely to respond best to.
Recommenders overview
Recommenders allow you to use any event properties from events in your datasets to predict which product, service, or other offer any individual will respond best to. For example, a recommender that uses the event property product will output the top products that an individual is likely to respond best to. This insight can help you tailor how you engage with them in order to achieve the best results. Recommenders aren't exclusive to just product recommendations–they can also be used to predict next action in a series, like next logical banking action.
Once you've created a recommender, you'll add it to the payload in a pipeline, then choose the cohort that you'd like to target with your recommender predictions. From that pipeline, you'll create a deployment to send your predictions to either a connection you've created, or to a CSV file.
Recommenders are just one of the ways that customer predictions can help you make every experience that a customer or lead has with your business exceptional. Try them in combination with persona predictions to maximize the customer experience.
📘When should I use a recommender instead of an outcome?
Recommenders are best suited to predict events that are part of a series, such as when you're interested in seeing if individuals will purchase product B or C after having purchased product A, or if the next logical action a banking customer might take is set up auto-deposit after having created an account. For conversion predictions, such as likelihood to churn, create an outcome instead.
Deleting a recommender
Before deleting a recommender, ensure that any resources using it, such as a pipeline, have first been deleted. Once there are no pipelines using the recommender, you can safely delete it.
- Dashboard: click the options menu (three dots) on the far right of the recommender you'd like to delete, or upper right when viewing a pipeline, then click delete.
- API: use the delete recommender API request.
📘Deleting resources
See object preservation for info on the order in which resources should be deleted.