The world of online shopping is actively integrating artificial intelligence: major players such as OpenAI, Google, and Amazon are investing in assistants that explore new product categories and suggest what to buy. Along with them, startups like Perplexity, Daydream, and Cherry are also building a business around discovering products with AI. In this context, Onton – formerly known as Deft – serves as an AI-powered platform for furniture shopping, which is already showing significant growth in active users and user engagement.
According to Onton, the active user base has grown from 50,000 to over 2 million monthly visitors, with millions of searches and image generations taking place.
The company announced it has raised $7.5 million in a new funding round, led by Footwork with participation from Liquid 2, Parable Ventures, 43, and others. Onton’s total funding to date is now around $10 million.
Onton co-founders are Zach Hudson and Alex Gunnarson.
Onton’s expansion plans and technologies
With the new round, Onton plans to expand into new categories, including clothing, and later consumer electronics. Previously the company transitioned from Deft to Onton due to confusion with the previous name and difficulties obtaining a premium domain.
Suppose you are looking for pet-friendly furniture. Our tools know that if a product contains polyester, it will be more stain- and scratch-resistant, and thus more pet-friendly. Our tools learn from every search and become smarter faster,
The founder explains: while large language models are good at guessing user intent, many tasks in e-commerce still require solutions. He adds that the average time a consumer needs to make a purchase decision is increasing.
For its core technology, the company uses a so-called neuro-symbolic architecture. According to Hudson, this approach helps avoid the hallucinations of large language models and deliver more logical search results. It is also emphasized that the model can learn from real-world data that is not always reflected in product descriptions.
Onton added various input methods and features to help people make decisions quickly and thoughtfully. You can upload an image or add a prompt, and Onton will find suitable furniture options for your home or office. There is also an endless canvas with image generation where you can combine existing images with found products for ideation. You can also upload a photo of your room and ask the tool to furnish it.
The company believes that instead of a single chat, these capabilities give shoppers greater freedom of action, even if they cannot precisely phrase a query in words. According to Onton, these approaches are already driving conversions several times higher than those of traditional online retailers, as users trust the data underpinning the services.
Hudson noted that thanks to technological and interface changes, launching a new clothing category will be easier. The company is building a catalog for this category and plans to launch the vertical in the near term. In the future Onton may face competition from Daydream, Aesthetic, and Style.ai.
According to Onton, the time from search to purchase is shortening, and the assortment continues to expand: the company currently has 10 employees with plans to grow to 15 by hiring engineers and researchers.
