Category ranking with ABR
This guide explains how Bloomreach Advanced Base Ranker (ABR) ranks products on category pages. It covers the signals the algorithm uses, how they're grouped, and how ranking scores combine to order your results.
NoteThis guide covers category ranking with ABR. For keyword search, read Search ranking with ABR. Category ABR is a separate algorithm from keyword ABR, so enabling one doesn't enable the other.
How category ranking works
A category page has no search query. Instead of matching a search term, ABR ranks the products in a category using category-browsing behavior and product performance.
ABR scores each product using key signals, then orders products from the highest score to the lowest. ABR learns how much each signal matters from your catalog's own behavioral data, so the ranking reflects how your shoppers browse and buy within your categories.
ABR organizes category ranking signals into 2 groups:
-
Category performance: how a product performs relative to the other products shown in the same category.
-
Overall product performance: how a product performs across your whole store.
Category ranking doesn't include a relevance group. Ranking on a category page is driven by category-browsing behavior and product performance, not by how well a product's text matches a search term. Because there's no search term, the algorithm doesn't use text relevance as part of the ranking score.
If a relevance score appears for a category result in Product grid insights, treat it as diagnostic information only. It doesn't contribute to the category ranking order.
NoteIf your catalog doesn't have enough category data for a catalog-specific algorithm, ABR uses the fallback Bloomreach global category algorithm instead. This algorithm includes only the 4 share signals and 4 sitewide count signals mentioned below. It doesn't include rate signals or custom signals, and you can't customize its weights. It can still include segmented signals if you have Real-time segments (RTS) configured.
Category performance
These signals measure how well a product performs on a specific category page. ABR uses proportions of activity (shares) rather than raw totals, which keeps rankings accurate whether a category gets 10 or 10,000 visits.
Share signals
| Signal | What it measures |
| Category product revenue share | The share of a category's total revenue that comes from this product. |
| Category product conversion share | The share of a category's total purchases attributed to this product. |
| Category product add-to-cart (ATC) share | The share of a category's total add-to-cart events attributed to this product. |
| Category product views share | The share of a category's total product views attributed to this product. |
Rate signals
Rate signals measure a product's efficiency or value within the category; a per-visit or per-order outcome, rather than a share of total category activity.
| Signal | What it measures |
| Category product revenue per visit (RPV) | Revenue generated per category visit associated with this product. |
| Category product average order value (AOV) | The average order value for orders containing this product, when the shopper found it by browsing this category. |
| Category product conversion rate (CR) | The conversion rate for this product among shoppers browsing the category. |
| Category product ATC rate | The rate at which shoppers browsing the category add this product to their cart. |
Overall product performance
These signals measure a product's popularity across your entire store, giving ABR broader context alongside how the product performs within the current category.
Unlike keyword search, category ranking doesn't adjust these signals by query relevance. There's no search term on a category page, so ABR uses these sitewide signals directly.
Count signals
| Signal | What it measures |
| Sitewide revenue | Total revenue this product generates across your store. |
| Sitewide conversion | Total purchases for this product across your store. |
| Sitewide ATC | Total add-to-cart events for this product across your store. |
| Sitewide views | Total views for this product across your store. |
Rate signals
| Signal | What it measures |
| Sitewide revenue per visit | This product's overall revenue per visit across your store. |
| Sitewide average order value | This product's average order value across all purchases in your store. |
| Sitewide conversion rate | This product's overall conversion rate across your store. |
| Sitewide ATC rate | This product's overall add-to-cart rate across your store. |
Product collections and sets
When a category contains product collections or sets, ABR ranks each set using the aggregated performance of its member products. A set with strong member-product activity ranks on that combined performance rather than being treated as a zero-performance product.
New products and new collections don't receive an artificial low score while their performance data builds up.
Segmented signals
If you have an active, supported Real-time segments (RTS) configuration, category ranking also uses segmented versions of the signals above. Segmented signals reflect shopper behavior within the active segment, alongside your store's broader unsegmented behavior.
Segmented category performance
-
Segmented category product revenue share
-
Segmented category product conversion share
-
Segmented category product ATC share
-
Segmented category product views share
Segmented overall product performance
-
Segmented sitewide revenue
-
Segmented sitewide conversion
-
Segmented sitewide ATC
-
Segmented sitewide views
If you switch between a segmented and non-segmented setup, re-train your category algorithm to reflect the change (this doesn't apply automatically). Read more about Real-time segments.
Override ranking
ABR trains on your catalog's behavioral data and learns the balance of signal weights that fit how your shoppers browse your categories. The default calibration is already tuned to your store. You can still override the ranking with:
Merchandising rules
Adjust the default order with boost and bury rules in the application, which continue to apply on top of ABR. Review existing rules if your signal weights change significantly.
Custom signals
Custom signals bring your own business data into the algorithm — for example, sales forecasts, margin scores, return rates, or stock levels. ABR trains these alongside its built-in signals and learns how much weight each one deserves.
For guidance on choosing and formatting custom signals, read the custom-signals best practices guide.
Updated about 1 hour ago