At Onton, we’re building the future of product search and discovery. It takes the average shopper 79 days to make a >$50 purchase decision, and we’re taking that to less than 1.

Our vision is to make e-commerce search as accurate as talking to an expert and as easy as asking for a recommendation from a friend.

We're a small team of 4 engineers, but here are some reads/listens that can help you learn more about us:

• Overcoming search challenges in e-commerce:
  <https://podcasts.apple.com/us/podcast/overcoming-search-challenges-in-e-commerce-with-alex/id1558589105?i=1000671499447>

• Competition and iterating:
  <https://podcasts.apple.com/us/podcast/ep-2-zach-hudson-onton/id1761011505?i=1000665167966>

• Why the name "Onton"?:
  <https://onton.com/blog/deft-is-now-onton>

The internet is messy. Years of search engine optimization (SEO) and misaligned incentives (e.g. advertising) have made information retrieval difficult and the output increasingly untrustworthy.

This is especially true in e-commerce, where entire categories of searches have been so over-SEO’d and over-advertised that searching them on Google and Amazon is essentially useless. As a result, today the average shopper spends 15 hours, across 12 different websites, over the course of 79 days trying to make a purchase decision.

Simply put, big search wasn’t made for online shopping.

And it’s only getting worse. Over the past four years, the amount of time it takes for someone to make a purchase decision online has increased by 9 days.

For all the impressive feats of large language models (LLMs), they end up compounding the pain points of e-commerce. LLMs are generating exabytes of unstructured, unverified data daily, and the tools at shoppers’ disposal today are neither prepared nor incentivized to deal with it. The burden then falls on the customer to separate fact from fiction.

Some platforms have begun to plug in LLMs like GPT and Llama to try to fix the problem, but the computer science adage of “garbage in, garbage out” applies: you can’t shortcut your way to better search by adding an intuition layer on top of bad data.

Merely organizing the world’s information isn’t enough to deal with this paradigm shift, and the advertising model doesn’t incentivize the changes that need to be made. Search engines need a revolution.

Our mission at Deft is to lead that revolution. We’re parsing the web to create trustworthy search, and we’ve started with the most troubled search category: e-commerce**.**