Is AI The Future of Shopping?
It’s no secret that the artificial intelligence (AI) revolution is upon us; a recent study found that 48 percent of CFOs plan to leverage automation and AI in 2024, while another study saw 44 percent of consumers say they felt that AI had helped retailers and brands create a more personalized shopping experience. Overall, AI in the retail market is valued at $9.65 billion, according to Statista—and is expected to hit nearly $39 billion by 2029.
But the speed at which AI is growing means that brands are pressed to stay ahead of the curve to meet consumers’ rising expectations.
More from Sourcing Journal
To that end, the South by Southwest (SXSW) festival in Austin, Tex., largely centered around AI, with more than 200 events and activations highlighting the technology. One panel, “How AI Will Change the Way Brands Sell and People Shop,” saw a line wrap around the JW Marriott and hit capacity (250 seats) over an hour in advance. Helmed by Google’s Lilian Rincon and J.Crew Group’s Danielle Schmelkin, the discussion focused on what brands need to know to keep up with cutting-edge technologies to embrace faster and more innovative ways to market and sell their products to a new generation of consumers.
And Google, being an “ecosystem,” sees billions of users a day come to the search engine to shop for a myriad of products on “all sorts of different journeys with different expectations,” explained Rincon, lead product of Google Shopping, to the packed room on the first day of SXSW.
“We’re really thinking a lot about personalization tailoring the experience,” she continued. So last year, the giant launched Google Virtual Try-On, which uses generative AI to show shoppers clothes on a variety of “real models,” comprising 80 different men and women of “all sorts of shapes and sizes.”
“The interesting thing about [Google Virtual Try-On] is that we decided to go with real people,” Rincon said. “A user chooses a model, and we can actually superimpose what that product would look like on the real model. And we’re really seeing people really get it.”
The team even got that technology to work with lay flat products, Rincon announced, which is ideal for merchants with smaller budgets who can’t afford a model shoot.
“We’re now able to actually take that [lay flat] and imagine what it would look like on a real human,” Rincon said. “Which is pretty remarkable.”
For J.Crew, personalization began with how consumers search for new products.
“We all search differently; even if we’re looking for the same thing, we may not be thinking about it in the same language,” Schmelkin, chief information officer at J.Crew Group, said, referring to the data attributes assigned to each product description. While those products previously had three or four data attributes, but the group “really wants to augment” every product with nearly 90 attributes full of language, synonyms and occasions relevant to that garment.
“This helps us tremendously—it helps with search, it helps our filtering, it helps with SEO; someone on Google could get to that shirt faster,” she continued. And we want to make sure everyone has that opportunity. I have a 16-year-old daughter, and she doesn’t search the way I search; she thinks very differently; Gen Z, they’re just different.”
Google has done “extensive research” with a bunch of Gen Zers and found that many of them are “much more open” to experiences with AI. When asked about shopping, Rincon said, Gen Z views it as a form of entertainment, with 20 percent of users needing to type five or more words to find the product they’re looking for. Google just launched a tool, SGE, in Google Labs that generates images based on those five words—also known as a prompt—that are then shoppable.
“And then we make those [images] shoppable, and if it isn’t exactly what you wanted, we have an edit box so you can continue, essentially, generating images to get that vision that’s in your mind into a product,” Rincon said. “We’re seeing good usage of it and people actually understand how to use it.”
While J.Crew is considering Gen Z as its next customer, it’s more focused on serving its current clientele, which skews older and may behave like Gen Z.
“What we want to make sure is, however you want to shop, we have an experience for you. You want to go to our website and do all your shopping online using the search bar, we want to make that efficient. You’re in a browser, we want to make that efficient,” Schmelkin said. “But we also want to make it connected and make sure it doesn’t feel like you’re having a different experience across all of those different channels.”
Google’s omnichannel experience has changed quite a bit over the last three years. In what Rincon calls “phygital,” Google has seen 60 percent of customers look up a product to find out that it’s in the store before they go buy it.
“There’s more of a blending in the way that people want to search,” Rincon said, adding that Google Lens has become increasingly popular, with over 12 billion monthly queries. “One thing that we recently announced is really about this convenience of this multi-modality, it was text before, it’s Lens now . . . we’re just evolving the way in which people are searching.”
Once consumers have found what they’re looking for, J.Crew is focused on making sure that customers can style it, creating the same experience online as having a shopping assistant in-store.
“For every product that we put up on our sites, we want to give three or four styling options [with] AI generated outfit recommendations,” Schmelkin said. “We’re also launching on-demand connect to a live stylist [because] sometimes you’ll find that it’s just not quite cutting through, and I actually want to talk to somebody. We really want to make sure we’re creating as much of that [in-store] experience as possible.”
