You are here:
Custom Objective for Offers
Custom objectives use deep learning models to predict the offers that are most likely to drive your business outcomes, even when customers don't directly interact with the offer itself.
How Offer Recommendations Learn from Indirect Signals
When you create a custom objective for an offer, the system uses machine learning to predict which recommendations are most likely to drive a desired goal, such as a purchase or a form submission.
To make these predictions, the model evaluates engagement signals such as clicks, product browses, referrals, and cart additions. It learns from these behaviors to identify which offers are most effective.
You can configure custom objectives for offers in two ways.
- Direct attribution:
- Optimize a direct action. For example, the system recommends a specific product and tracks whether the customer buys that same product.
- Indirect attribution:
- Optimize a downstream action tied to an offer. For example, the system recommends a "10% off" offer and tracks a product purchase where that offer is applied.
- Create a Custom Objective for Offers
Using indirect engagement signals, create custom objectives for offers in objective-based recommenders. This objective trains the machine learning model to deliver relevant, personalized offers that drive your business goals. To do this, the model links promotional offers to downstream conversions, such as product purchases.
Did this article solve your issue?
Let us know so we can improve!
