Shaped Launches Semantic Search with Behavioral Re-ranking

The team at Shaped is thrilled to announce the launch of semantic search with behavioral reranking. In our new, unified system, Shaped’s text retrieval models work hand-in-hand with behavior-driven re-rankers to give your users industry-leading search personalization in less than one sprint.

Why search?

At Shaped, our mission is to help our customers easily harness behavioral data to enable their users to find exactly what they’re looking for. With digital content exponentially increasing from generative media, the creator economy, and live applications, the challenge of content discovery at scale has never been greater. Your catalog of content and products is only as powerful as your users ability to discover it, and Shaped is built to cut through the noise.

Where does Shaped fit in the search landscape?

You might wonder: with mature search solutions like Algolia, Elastic, and a slew of emerging vector-based offerings, do we really need another player? The answer is yes. Existing solutions assume your users have uniform expectations for their search results. In reality all users are unique, with nuanced intent. Sure, existing systems handle text search and semantic queries well. But when it comes to delivering results that actually drive business impact, integrating behavioral data is the secret sauce.

Take this example: Suppose you're an e-commerce site, and a user searches for "linen pants." A traditional semantic search might return a generic assortment of styles for both men and women. But what if that user previously clicked on the "menswear" category or purchased men's items? Suddenly, it’s obvious: the results should prioritize "men's linen pants," even if the user didn't explicitly specify it. This is where Shaped shines—using behavior-driven insights to deliver relevant, personalized search results, all based on the unique context of each session.

Personalized search: Not just for Big Tech anymore

Behavior-driven search has become standard at companies like Google, where your search results differ from the next person because the system understands you as an individual. It’s not just convenient; it’s incredibly lucrative. The catch is, these companies spray multi-million dollar budgets over year-scale timelines to achieve these outcomes. Not only is Shaped’s semantic search easy to integrate, it has big-tech levels of customizability, with the ability to fine tune your models at your fingertips.

In our own experience, businesses that employ personalized semantic search see conversion rates increase by more than 10% from baseline—and that's not even counting the long-term value of delivering a more enjoyable, relevant user experience.

What this means for you

With Shaped you can now implement big-tech-levels of search and recommendations without multi-million dollar budgets or lengthy development timelines. Integrate, experiment, and evaluate with behavior-based semantic search and recommendations in one place, and watch your engagement, conversions and revenue soar.

We can’t wait to see what you build. Click here to try 7 days of Shaped for free.

Get up and running with one engineer in one sprint

Guaranteed lift within your first 30 days or your money back

100M+
Users and items
1000+
Queries per second
1B+
Requests

Related Posts

Tullie Murrell
 | 
December 5, 2022

Real-time Search, Session, and Similar Ranking

Tullie Murrell
 | 
March 22, 2023

Shaped 1.0: The fastest way to personalize your product, platform or marketplace

Jaime Ferrando Huertas
 | 
March 14, 2023

GPT-4: A New Milestone in Scaling Up Deep Learning