Predictive Purchases
Knowing which of your users is likely to make a purchase is a crucial insight for growing businesses. Without it, how do you decide which campaigns to build? Who should receive discounts and promotions? Where to spend a limited budget? Braze helps answer these questions with Predictive Purchases, a machine learning model that makes it easy for marketing teams to understand future purchasing behavior and focus their resources on revenue-maximizing campaigns.
Topics
Overview
Predictive Purchases give marketers a powerful tool for identifying and messaging users based on their likelihood to make a purchase. When you create a Purchase Prediction, Braze trains a machine learning model using gradient boosted decision trees to learn from previous purchase activity and predict future purchase activity.
Once a Prediction is built, users are assigned a Purchase Likelihood Score between 0 and 100 denoting how likely they are to make a purchase. The higher the score, the more likely a user is to make a purchase. Users are also sorted by Low, Medium, and High Purchase Likelihood Categories.
The real value of Predictive Purchases lies using Prediction results to create a Segment or campaign. Marketers can build targeted campaigns directly on the Prediction page for immediate revenue-boosting results or save a Segment for a future campaign or Canvas. Not sure who to target first? Read our strategic considerations for messaging users based on their Purchase Likelihood.
Access Predictive Purchases
The Predictions page is located in the Analytics section. For full access, contact your account manager.
If you are using the older navigation, you can find Predictions under Engagement.
Prior to purchasing this feature, it is available in Preview mode. This will allow you to see a Demo Purchase Prediction with synthetic data as well as create one Preview Purchase Prediction model at a time. This Prediction will be created based on your actual user data, but it will not allow you to target users for messaging according to Purchase Likelihood. It will also not update regularly after creation.
With the Preview, you can also edit and rebuild this one Prediction or archive it and create others to test the expected Prediction Quality of different audiences and become familiar with the analytics.