Bloomreach Search is an AI-powered site search solution that drives revenue by delivering the most relevant results as per your shoppers’ intent. Bloomreach Search is built to

• Interpret a variety of complex natural language queries/expressions
• Find product results as per your shopper's product search intent

What sets Bloomreach Search apart

---

Semantic Understanding

Bloomreach Search leverages Semantic Understanding to interpret the context behind each query. Semantic search uses natural language processing to detect matches on the query intent, while traditional Keyword-based search only finds literal word matches.

Let’s understand this with an example:
A Keyword-based search engine cannot tell the difference between “dress shirt” and “shirt dress," since it lacks actual query understanding. This often results in poor search relevance.

Our semantic-powered algorithms know that “dress shirt” and “shirt dress” are different products. They recognize this difference by identifying product types and attributes on both the user query and product content.

Semantic understanding, therefore, helps return results that are more relevant to the user's needs, even when the query is phrased differently.

📘

[Optional] Read this article to learn how Semantic Search enhances product discovery experiences.

Self-learning & Evolving Search Intelligence

Bloomreach Search is always learning from the shoppers’ site engagement behavior. Our system has been trained on a dataset that we’ve built over the past decade to best adapt to customer and buyer behavior.

Bloomreach Search comes with zero-day learnings that equip your search box with our decade-long commerce experience. With continual learning, Bloomreach Search gets smarter each day and adapts rapidly to your customers’ ever-changing needs as you scale.

How Search Works

---

Bloomreach’s Search provides an optimal search experience by nailing two important functions:

1. Recall

Our recall algorithms find the set of products in your catalog that match the visitor’s search query. This is achieved by:

• Extracting product attributes
• Interpreting product intent with semantic understanding
• Matching all query parts
• Enhancing the query with Synonyms
• Handling challenging queries like typos, assortment gaps, numeric queries, and long-tail queries to retrieve precise recall

2. Ranking

Our commerce-specific algorithms arrange the recalled product set in an order that is optimized for KPIs like RPV and Conversion rate. Performance signals, relevance signals, and several important factors are combined to assign a score to each product. These scores are then used by the algorithms to automatically rank the products in a way that generates the most revenue for your business.

This combination of precise recall and optimized product ranking guarantees that your shoppers have the most relevant browsing experience while also maximizing your revenue.

📘

Guided Search Onboarding

If you are looking for a guided onboarding on Bloomreach Search, visit the Get to Know Search article.