Algolia is excited to announce the release of Query Categorization, an all-new AI-powered feature that automatically improves search results for all your categories, producing higher conversions and happier customers.

Query Categorization was designed for product managers and merchandisers looking for new ways to improve on-site performance. How does Query Categorization help? It automates much of the complexity of searchandising optimization. 

Before Query Categorization, product and category managers would have to create endless rules  to improve on-site search results for each category. This was particularly painful for large catalogs with thousands of SKUs. Often it would require enlisting the support of data scientists and experts to assist them with enriching the data.

Query Categorization can eliminate much of the complexity and time required to classify products. Query Categorization uses an AI model to automatically surface product categories associated with your users’ individual queries. It predicts the categories to which a search query belongs. It allows you to automate manual processes, unlocks automatic boosting & burying, delivers higher-level analytics, and promotes more efficient merchandising.

How it works

We can use an example of an online grocery shop to see how it works. A customer searching for products in the shop might type in a query for “banana.” In this case, Query Categorization will automatically include it as part of the category Food > Vegetables and Fruits, boosting actual bananas above products that have bananas as an ingredient. The new categorization improves end-user search results — and eliminates the effort that would have normally been required.

Algolia’s Query Categorization includes:

• A dedicated section in the Algolia UI (user interface) to set up AI models and explore  predictions in an easy and intuitive fashion;
• Automatic filtering and boosting on predicted categories using a no-code environment  that business users and non-tech practitioners can control directly to increase the relevance of their user’s results;
• Full control is retained by merchandisers over Query Categorization, and they can modify any categorizations made by the AI model;
• Analytics available in the UI that are grouped by predicted category to learn how the categories perform and to detect underperforming queries;
• Access to category predictions at query time (with the Search API) so that a personalized Search and Discovery experience can be provided for consumers.

Early testers have been thrilled with their ‘boosted’ results — up to a 22% higher conversion rates. A major European marketplace has seen significant enhancements to both its click-through rates (CTR) and conversion rates (CVR). 

Let’s get started!

Query Categorization is available to Premium customers. To start, you’ll need to set up and send click and conversion data to Algolia. Second, specify how you want to classify your records in the Query Categorization section of the Algolia dashboard — no is coding required. Third, sit back and let the model automatically learn how to categorize your end-users’ queries. It’s easy to see what predictions the system is making before you turn it on for your users. Give it a shot today!

Current Algolia customers can learn more by visiting the Algolia Academy.

Read the full press release.