REVIEWS.io review data now drives smarter product discovery in Boost

We’ve expanded the REVIEWS.io and Boost Commerce integration to allow review data to play a more active role across search, filtering, and merchandising. Merchants can now use ratings and structured review attributes to influence how products are displayed, filtered, and promoted throughout the storefront.

What’s now available

Reviewer-based dynamic filters

Boost can now turn structured REVIEWS.io review attributes into dynamic storefront filters. Shoppers can filter products based on real customer experiences collected through review attributes, such as fit or occasion.

Examples include:

• Fit: true to size, runs small, runs large
• Occasion: wedding, party, everyday use
  

This feature is available to merchants on the REVIEWS.io Grow plan or higher. Text-based attributes are not supported for Boost filters.

Rating-based merchandising rules

Merchants can now use REVIEWS.io ratings as a product attribute when setting up Boost merchandising strategies. This allows ratings to influence how products are boosted, demoted, hidden, or filtered.

Rating-based rules can be used to:

• Promote high-rated products
• Downplay or hide low-rated items
• Combine ratings with other merchandising rules
• Run A/B tests to measure the impact of review-based strategies

REVIEWS.io rating attributes can also be used with Merchandising by Markets.

Re-ranking of Boost search and recommendation results using REVIEWS.io data

Boost can now apply REVIEWS.io review data directly within its search and recommendation models, allowing review signals to contribute to how products are ranked and surfaced during discovery.

When enabled, Boost considers multiple REVIEWS.io data points, including:

• Rating score (1–5 stars): higher-rated products are prioritised
• Review count: products with a greater volume of positive reviews gain visibility
• Sentiment: products with more positive review sentiment are promoted more strongly
  

These review signals are applied alongside Boost’s existing ranking logic, which prioritises:

1. Relevance
2. Product performance metrics such as conversion rate, add-to-cart rate, and click-through rate
3. Review data, with sentiment weighted more heavily than star rating
   

Setup is handled within Boost. When this setting is enabled for the first time, Boost will take approximately 30 minutes to train on the review data before changes are reflected on the storefront.

‍View the integration setup guide.