Client: Industrial Supplies Retailer
A leading supplier of residential and commercial construction supplies wanted to increase ecommerce revenue by personalizing product recommendations. Previously, they produced generic product recommendations in high-traffic areas on their website. They showcased top-selling products and new items, but they weren’t seeing a substantial return.
Sophelle first analyzed their AI algorithms. Because the retailer prioritized generic product recommendations prizing top selling and new products, we needed to alter the weight of that strategy to improve performance. To increase revenue, new data models deemphasized their general approaches in favor of AI algorithms that delivered more relevant product recommendations based on consumer behaviors. While customers are often inclined to purchase top sellers, this client’s customers were much more likely to buy items that other customers have actually browsed through.
We then ran A/B tests to analyze the results. Tests revealed that targeted recommendations using consumer behavior data drove substantially higher conversion rates than browsing data.
By adjusting the algorithm to leverage customer-specific, real-time recommendations, Sophelle’s client achieved a 265% lift in personalization revenue.