Query understaNding

Translate customer queries to business data

Sizes, shapes, colors, materials, years, makes, models, etc! Queries often include structure with very specific intent. Language can also be very vague and up to 25% of queries may be misspelled. Fortunately, Search.io’s powerful query understanding models ensure that your customers can type just about anything into your search box and get great results.

search spelling and NLP

Natural language processing

Built-in language processing capabilities ensure great results, even when your visitors misspell words (like we did above), query on broad categories or exact product names, or enter long multi-word queries.


Spelling mistakes happen to the best of us; don't let them ruin your customer's search experience. "Gold shoes" may mean "golf shoes," but do you really know? Search.io uses probabilistic models to analyze your records and clickstream data to best anticipate the user intent.


Queries such as “returns” and “store near me” won't ever match any products, but they're critical to customer flow. Redirects allow you to bypass running the query and send the user directly to the right URL.


Customers may refuse to use your fancy product name and call it something else. No problem. You can rewrite the query, find and replace, regex extract, and much more.


Neuralsearch understands languages and removes the need for all common synonyms — but you can also manually add synonyms for your proprietary products and services if needed.

multi-language search

Multi-language support

Nosotros hablamos español… and Deutsche, 日本語, and français. Language packs available for multi-lingual search. We support Latin script languages and multi-byte symbol-based queries, such as Chinese and Japanese for keyword search. Neuralsearch natively supports 16 languages for more advanced AI understanding.

query and index pipelinessearch result sorting


Pipelines allow you to configure data ingestion and search algorithms via simple YAML-based notation. This allows you to store your entire config-as-code. Pipelines give you flexibility to chain together series of steps to execute complex instructions to both define how your data is ingested and also to build and process your queries.


Do you want a pretty search solution or a highly configurable one? That’s how most search-as-a-service solutions are offered. We didn’t like that tradeoff. Pipelines give you the power to do just about anything — clone the config, audit tiny changes, view diffs between versions, or add powerful processing… all in just a few lines of markup.


Changing a setting is simple. How about triggering complex logic that can be composed to combine amazingly powerful search? If you love GitHub Actions you will love pipelines.

Enrich data with third-party services

You want to call an external service. Go for it. Pipelines can call out to third party services to process information. In practice this is used to turn images into colors and descriptions, classification, clustering, custom dynamic pricing, and more. For example, you can call out to an external vision API to extract metadata from an image so you can filter on that data.

Version control

Every pipeline change creates a new version. Why? Because you can move forward and back in time at will. Every group of changes can be A/B tested, unwound as needed. You can even do a code diff inline if needed to see exactly what changed.

Introduce intelligent search to your site seamlessly in minutes.