Item Recommendation Model for Failed Purchase Re-Engagement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems fail to remind customers about out-of-stock items they were previously interested in purchasing, leading to a high chance of missed opportunities for re-purchasing these items.
Innovation Solution
A machine learned model tracks unavailable items from previous shopping lists and predicts customer intent to re-purchase using an intent prediction model, generating recommendations when the items become available.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional systems substitute or refund unavailable items, then order fulfillment is maintained, but customer satisfaction decreases and re-purchase opportunities are lost
Solution Approach 1:
The system performs preliminary actions by tracking unavailable items during the first session and preparing recommendations in advance. When the item becomes available in the second session, the system proactively generates and presents recommendations before the customer needs to repurchase, thus preventing satisfaction loss and capturing re-purchase opportunities.
2Productivity
If the system tracks and remembers unavailable items, then re-purchase opportunities increase, but system complexity increases
Solution Approach 1:
The system segments the shopping interaction into distinct sessions: a first session for initial purchase attempts and a second session for follow-up recommendations. This segmentation allows the system to track unavailable items specifically without overwhelming complexity, focusing resources on the critical re-purchase moment when items become available.
Solution Approach 2:
The system implements feedback loops by monitoring whether tracked items are repurchased in the second session. This feedback mechanism allows the system to learn from customer behavior patterns and refine its recommendations, improving re-purchase conversion while managing complexity through data-driven optimization rather than overly complex architecture.
3Measurement precision
If the system uses machine learning to predict intent, then recommendation accuracy improves, but computational resources increase
Solution Approach 1:
The system applies partial action by using machine learning intent prediction only for tracked unavailable items rather than all products. This selective application reduces computational resources significantly while maintaining high recommendation accuracy for the critical items that need repurchasing, avoiding the excessive computational cost of analyzing every possible product recommendation.
Data Source
AI summary
A machine learned model for item recommendations following failed attempts to purchase those items. During a session, an online system receives a request to fulfill an order from a user device. The system receives a message indicating that an item from the order was not fulfilled. The system logs the item in connection with a profile of the user stored in a database of the online system. During a subsequent session with the user device, the system determines that the logged item is available for fulfillment. The system applies the model to output an intent score indicative of an intent of a user of the user device to acquire the logged item. The logged item is ranked based on the intent score, and a user interface is generated that includes a recommendation to acquire the logged item. The system causes the user device to display the generated user interface.