This post was originally published on this site.
The first age of online personal shopping tools came via browser toolbars and other download services, such as Honey, that tracked deliveries and scoured for coupons.
The new thing in personal shopping tools is, of course, AI agents.
But most shopping AI agents are firmly attached to one particular LLM or retailer, like ChatGPT and Perplexity’s new shopping response flows that guide their own shopping-related queries. And then there’s agentic shopper pups: Walmart’s Sparky and Amazon’s Rufus, which are meant to help customers shop across their retail footprints.
What people need is an independent agent that peers across retailers and is entirely focused on ecommerce services. At least that’s the conclusion driving Wizard Commerce, a personal shopping agent that emerged from beta on Wednesday.
The biggest LLMs and retailers have a huge built-in user base for their shopping agents. But “distribution doesn’t create a great product,” said Wizard AI Co-Founder and CEO Melissa Bridgeford. She started the company with Marc Lore, CEO of the Wonder Group and former CEO of Walmart US Ecommerce.
Google has had the world’s best distribution and canvas for a personalized shopping service but has not delivered much in the way of any shopping experiences, Bridgeford said. She cited services like Zillow or Kayak that are superior within their specialties (real estate listings and travel searches, respectively), as Wizard hopefully will be for ecommerce.
Wizard Commerce isn’t putting itself in the position to grow an app business from zero, with all the costs that come with new user acquisition. To start, the service begins with a URL-based search engine that’s free for people who sign up with an account tied a phone or email.
From there, a shopping query will return follow-up prompts (e.g., a search for clothes might be narrowed down by style or size) until five items are returned in a product carousel. None of these is a sponsored listing, Bridgeford said, nor can a brand or retailer pay to influence the results.
Right now, Wizard’s results all include a button that sends the shopper to a retailer page to pay. Bridgeford said Best Buy is the first retailer that Wizard is working with on a native checkout functionality – so users can purchase directly within the Wizard feed, rather than link out.
Ironing out these retail integrations is key for Wizard, considering the product is free and the stated plans are to never accept ads. The goal will be for Wizard to earn a rate or fee based on the sales it will generate.
“Business model is probably not something we want to comment on right now,” Bridgeford said.
Subscribe
AdExchanger Daily
Get our editors’ roundup delivered to your inbox every weekday.
Daily Roundup
The company hasn’t raised a funding round, either, though its co-founders are both experienced seed round investors themselves.
Wizard is also sticking to its consumer-facing business stance. For instance, there are many agentic commerce companies that operate chatbots for retailers and merchants (usually as an annoying pop-up). Bridgeford said Wizard is committed to sticking with the B2C product.
“I think the pain point is huge,” she said.
There’s been an explosion of ecommerce data not just from the perspective of ecommerce marketers. People are overwhelmed by product reviews on Reddit, unboxings on YouTube, reviews on social media and more places, often for basic things like clothes, soda or makeup. And then the shopper must navigate numerous marketplaces even once they have decided on a purchase.
This problem has existed for years, Bridgeford said. She and Lore started working five years ago on what would become Wizard Commerce, though at the time, she said, they were both considering investments in a startup category that was then called “conversational commerce.”
“Fast-forward to today,” she added, and those conversational commerce roots have grown into a fully fledged shopping agent.
Now it just has to work like magic.