The Granularity of Discovery: How SKU-Level Attribute Mapping Enhances Search Relevance and Consumer Conversion in ECommerce

Abstract

The article centers on how SKU-level attribute mapping can be relevant in improving relevancy and customer experience in online stores, which is the primary retail environment/shop, when dealing with art supplies. It points out that classical title, brand, price metadata search models are incapable of providing high-intention search results based on technical data such as pigment code, ASTM lightfastness, paper weight, texture, transparency, or even nib type. The methodology used in the study is based on a comparative audit of 25,000 SKUs, in which a control group (SKUs with standard metadata) and an experimental group (SKUs with enriched SKU characteristics) were used. The results show that granular attribute mapping increased the searchto-product click-through rate by 4.2% to 7.8%, the conversion rate by 1.9% to 3.4%, and the average order value by $64.50 to $ 82.10, and also decreased the number of zero-result queries from 21.5% to 5.1%. The study concludes that long-tail SKU attribute mapping is better for long-tail search and minimizes ambiguity around search, trust, and compliance, but it cannot be extended into an SEO tool; instead, it is a product-discovery infrastructure in modern online trading. It also suggests using metadata depth as a strategic approach to integrating Product Information Management, faceted navigation, search engineering, and revenue architecture. The research areas that have arisen include expanding SKU-level modular AI attribute mapping to large catalogs using adaptive taxonomy design, visual search, privacy-first personalization, and relevance, fairness, and transparency