Product Reorder Notification System Using Inter-Purchase Interval Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face inefficiencies in reordering online items as they often need to browse multiple pages to locate previously purchased products, which is time-consuming and prone to forgetting items, especially for frequently reordered products with long inter-purchase intervals.
Innovation Solution
A system that processes historical marketplace information to group products into clusters, calculates inter-purchase interval likelihood scores, identifies candidate products for reordering, determines optimal notification times, and ranks these products for display in a graphical user interface, facilitating reordering through a streamlined notification process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If users manually browse multiple pages to locate and reorder items, then they can find and reorder products, but the process becomes time-consuming and inefficient
Solution Approach 1:
The system performs preliminary actions by analyzing historical purchase data beforehand to identify frequently reordered items. It pre-calculates inter-purchase interval likelihood scores and prepares personalized reorder recommendations, so that when users need to reorder, the items are already identified and ready for quick selection, eliminating the need to manually browse multiple pages
Solution Approach 2:
The system enables self-service by automatically generating reorder recommendations based on user purchase history without requiring manual intervention. The system autonomously identifies candidate products, calculates likelihood scores, and presents personalized reorder lists, allowing users to simply review and confirm orders rather than actively searching for items
2Reliability
If users manually search for reorder items, then they can locate products, but they may forget some items they prefer to re-order
Solution Approach 1:
The system implements feedback by continuously monitoring user purchase history and inter-purchase intervals. It uses this feedback to dynamically update likelihood scores and refine reorder recommendations, ensuring that the system learns from past behavior and improves its accuracy in predicting which items users will want to reorder, thereby reducing forgotten items
Solution Approach 2:
The system performs preliminary analysis of purchase patterns to identify candidate reorder items before the user even thinks about reordering. By pre-processing historical data and calculating likelihood scores in advance, the system ensures that all potential reorder items are captured and presented to the user, eliminating reliance on user memory
3Quantity of substance
If the system displays all candidate products in a single list, then users see all options, but the interface becomes overwhelming and difficult to navigate
Solution Approach 1:
The system applies segmentation by dividing the candidate product list into multiple sections based on likelihood scores or product categories. High-probability reorder items are separated from lower-probability items, allowing users to focus on the most relevant options first while still having access to the complete list if needed, thus making the interface more manageable and easier to navigate
Data Source
AI summary
Systems and methods including one or more processors and one or more non-transitory computer readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform: receiving historical marketplace information for a user in a marketplace corresponding to products previously purchased by the user; processing the products to group the products into one or more product-type clusters; analyzing the one or more product-type clusters to determine respective inter-purchase interval (IPI) likelihood scores for each product in each of the one or more product-type clusters; identifying one or more candidate products from the one or more product-type clusters that have respective IPI likelihood scores that satisfy one or more thresholds; determining a respective time and a respective duration for a respective re-purchase notification for the user based on the respective IPI likelihood score for each of the one or more candidate products; ranking the one or more candidate products based on the respective IPI likelihood scores for the one or more candidate products; and transmitting a re-purchase notification to the user via a graphical user interface (GUI) that includes at least a subset of the one or more candidate products, the GUI including a first section that includes a first portion of the at least the subset of the one or more candidate products and a second section that includes a second portion of the at least the subset of the one or more candidate products. Other embodiments are disclosed.