Understand ranking with Variant slicing

This guide explains how Bloomreach calculates ranking scores for Variant slicing in search results.

How ranking works with Variant slicing

With Variant slicing, the Search ranking algorithm determines how product variants appear in results and which variant becomes the hero (that is, the primary variant displayed). 

Variant slicing promotes the most relevant and best-performing variant presentation for a given shopper query. It looks at how each variant has performed and how well it matches the query, then surfaces the strongest variant as the primary result. Then, it groups the rest as alternatives that shoppers can switch between.

The algorithm distributes a product's overall ranking score among its individual variants using a three-step process that ensures the most relevant and best-performing variants appear prominently in search results.

Step 1: Distribute the product ranking score

Every product starts with a total product ranking score (also called the performance score). Variants that convert better are treated as stronger candidates and receive a higher score.

Step 2: Add relevance scores

A variant that matches the query's specific attributes (like the color someone searched for) is treated as more relevant than one that doesn't.

Read on the importance of sending SKU events. 

Step 3: Select hero variants and form listings

Once each variant has a combined score (product score + relevance score), the algorithm groups variants into listings.

The variant that best matches the query and has the strongest track record becomes the hero variant, the version shown first. Other variants of the same product are grouped into the same listing as alternatives, so shoppers can still find the color or size they want even if it isn't the one shown by default.

Requirements for accurate ranking

To ensure Variant slicing ranking works correctly, send SKU IDs with your conversion events (add-to-cart, purchase, and other variant-level signals). Without SKU-level conversion data, the algorithm can't determine which specific variants are actually converting, and falls back to treating all variants equally.

If conversion events lack SKU IDs, the algorithm distributes the product ranking score equally among all variants. That produces a less accurate hero-variant selection and can hurt search performance. Configure your conversion tracking to include SKU IDs for all variant-level events to get the full benefit of Variant slicing.