Article

In the fall of 2022, I received an email from one of our most trusted technology partners, Algolia, announcing its acquisition of Search.io and the upcoming release of Algolia NeuralSearch, its hybrid search engine, currently in private beta.*  

Once released, Algolia will be the only search and discovery platform to include both keyword and semantic search in one API. This is great news for Algolia customers and why enterprise brands, ecommerce businesses, or anyone with a high-traffic, resource-rich site should consider using the platform.

But what is semantic search and why is their hybrid search engine so special? Let’s take a look.

Semantic vs Keyword Search

In a nutshell, semantic search more accurately reflects the way we think as humans and increasingly how we search. This is especially the case when we’re not exactly sure what we’re looking for. We search around the thing using descriptive words and even “long tail” phrases taking a more-is-more approach. This is in contrast to keyword searches which work best when we already know what we want. We enter a few specific words–direct and to the point. Let me give you an example.

Let’s say you see someone wearing a pair of cool Dr. Martens. You think, “I want boots like that!” so you go home and Google, “Dr. Marten boots.” You’re then rewarded with all kinds of relevant options and quickly find what you’re looking for.

But what if you didn’t know they were Dr. Martens? You might enter something like “cool black chunky leather rock and roll boots.” Within those search results, you will certainly find Dr. Marten boots. (Try it and see.) That’s because, with semantic search, a connection between Dr. Marten boots and “cool”, “chunky,” and “rock and roll” boots is understood.

Which leads us back to Search.io and NeuralSearch™, the semantic search solution that Aloglia acquired last fall. 

Semantic search understands connections and context, thanks to machine learning. NeuralSearch’s A.I. was trained on billions of documents allowing it to predict user intent and meaning in a way keyword search engines cannot. That kind of processing power requires a lot of resources which has traditionally made semantic search implementations cost-prohibitive for many businesses. 

But Search.io found a solution. It developed an innovative hashing technique to drastically reduce the resources required to compute, store, and retrieve vectors (which powers semantic search). That made semantic search scalable, cost-effective, and a perfect addition to the Algolia search and discovery platform.

The Benefits

By combining keyword and semantic search in a single API, Algolia customers will benefit in major ways. Here’s how:

• Easy, out-of-the-box setup.

  You’ll no longer need to add synonyms, complex search rules, keyword-stuff titles, or plug-in additional language packs when setting up your search implementation.

• Increased conversions.

  By combining keyword and semantic search, customers will experience the highest level of relevance and accuracy in their search results which in turn boosts sales and other conversions. 

• Improved user experience.

  Users will be able to enter keywords or write how they think using short or long-form expressions and natural language.

• Competitive advantage.

  No other vendor offers this two-in-one solution. That means Algolia customers will have an edge over competitors using alternative search and discovery solutions.

• Affordability.

  All of these benefits will be available at a price that’s 90% less than other semantic search solutions on the market today. 

Next Steps

Interested in hybrid search for your business? As mentioned, Algolia’s new product is currently in beta and will be available soon*. If you would like to stay in the loop about Algolia NeuralSearch, you can sign up for updates. If you would like to learn more generally about how Algolia and other best-in-class composable solutions can benefit your business, please contact us. We would love to chat.

*Algolia has released NeuralSearch since the launch of this blog post.