EC Recommendation Guide | From Mechanisms, Algorithms, Tools, and KPIs
"We want to add a recommendation function to our e-commerce site, but we don't know which method to choose." "We've implemented a recommendation engine, but we can't explain how it's impacting sales." "We often hear the term 'AI recommendation,' but we haven't fully grasped how it differs from traditional recommendation systems."
Introduction
On the other hand, if you introduce tools, algorithms, and KPIs without clearly defining them, you are likely to end up in a situation where "you can't see the effects of what you've implemented."
1. What is EC Recommend?
EC recommendation is a system that presents products tailored to each customer's interests on an e-commerce site. Based on browsing history, purchase history, attribute data, and behavioral data, it predicts and displays products that "this customer is likely to buy next" or "products they might be interested in."
1-1. Why is e-commerce recommendation important?
What's important here is that of the remaining 97-98 people, a significant proportion have been exposed to products they are interested in, even if they don't end up purchasing them. Recommendations act as a tool to present this group with their "next move," thereby boosting conversion rate (CVR), average audience (AOV), browsing rate, and repeat purchase rate.
1-2. Typical EC Recommendation Scenarios
Each scenario's role changes depending on its placement and the customer's purchasing phase. The basic approach is to design it by separating it into "what to display at the top," "how to guide users through product details," and "what to suggest in the shopping cart."




