Microscope can give you some insight into the best products for an individual user. The product sections are split into two distinct groupings.
- Last Purchased Items
- Product Recommendations
In this article we'll give you an understanding of how Cortex determines the product recommendation as well as where the data comes from to inform the last purchased items section. We'll begin with the last purchased items as they provide some more context for the product recommendations:
Last Purchased Items
This information is passed to us through your integration with Retention Science. Cortex displays whatever is considered a valid order and looks at the items that were purchased. This allows you to understand a bit more about the user and understand their interest in specific products. Cortex uses this to map users in terms of affinity and uses it to predict the users' next purchase. This is based on complimentary and cross selling trends that are captured in your historical files and current purchases from a site level. It also will help identify the product in the Follow Up, and replenishment stages. Furthermore, our Predicted Product Preferences segmentation is informed by recent and past items that have been purchased. Talk to your CSM or the support team on adding this feature to your account.
The recommended items that are displayed are the items that are most likely to convert a user. These recommended items are based on browsing behavior, order history and complimentary products. The entire items file is incorporated in the calculation of product recommendations so the Data Science can pick up on implicit and deep connections between products and users. One thing to note is that the product recommendations displayed in the email a user receives may not be the same as the recommendations shown in microscope. The reason is that each stage has a slightly different recommendation scheme that influences the recommendations for that specific stage. For more information on the differences in the specific stages, check out these articles.
That's the basics of the Predictive Analytics for Microscope! You can check out the descriptions on the other sections below: