AI-powered visual search is gaining rapid adoption among retailers, helping customers find the products that inspire them. Retailers that have adopted the technology report a significant increase in conversion rates and average order value. Meet the company making it a reality
Farfetch is an online retailer selling luxury menswear and womenswear, a sector with ferocious competition. So, to distinguish itself from rivals, Farfetch offers customers a radical new technology known as visual search.
It works like this. Customers use the Farfetch app to upload a photograph of an item of clothing. It could be a photo of a friend or a screen grab of a celebrity they’ve seen on Instagram. An artificial intelligence (AI) engine then decodes the image, identifying the colour, style and fabric of items of clothing in the frame. Then Farfetch publishes similar items for the customer to browse.
It’s quick and intuitive. Finding the shoe Emma Watson wore to the Oscars or something similar to that cute sarong your bestie wore to the beach is astonishingly simple.
Young customers adore it. It fits their way of shopping, putting images at the heart of the experience, natural for the Instagram generation. It also encourages spontaneous purchases. As Accenture found: “Gen Z consumers place a greater emphasis than millennials on turning to social media for inspiration. And nearly 60 per cent more say they made a purchase just because they wanted to buy something or because they randomly saw an item they liked.” Someone admiring a friend’s outfit can instantly find something similar online using only an image.
The arrival of visual search in retail is in a large part due to Syte, a visual AI company with offices in Tel Aviv, London and Geneva. “Syte connects consumers’ inspiration and the real world with retailers’ inventory, no translation needed,” says Lihi Pinto Fryman, Syte’s chief marketing officer and co-founder.
Syte provides visual search for some of the world’s largest retailers and marketplaces including Argos, Tommy Hilfiger, Marks & Spencer, Bonprix, PrettyLittleThing and many more. It is renowned for the high accuracy of its search results, speed, and ease of implementation, making it the leader in the field.
For a brand such as Marks & Spencer, partnering with Syte is an opportunity to help consumers take control of the way they discover products, enhancing their consumer experience.
Jim Cruickshank, head of digital product and user experience at Marks & Spencer, says: “We know our customers are busier than ever and are often most inspired when they’re out and about. Style Finder [a Syte-powered tool] helps customers instantly find what they’re looking for, without the need to manually search and filter through our products.”
Image search is just one of the ways Syte can help retailers. A parallel offering is the “view similar” function. Customers find an item they like in the retailer’s store, then click an icon to view a range of similar products also offered by the retailer. Some may be cheaper or offer a twist on the style. It’s a fun and practical way to explore an online store.
Product tagging can be done automatically by Syte. The traditional method is labour intensive, with manual operatives entering terms such as “polka dot”, “skirt”, “maxi”, “women’s”, “high waisted”, “satin” and so on. Syte can automate the entire tagging process. This slashes back-office costs. Retailers can do predictions and analytics. And consumers have an alternative way to search the store, increasing basket sizes and engagement.
Perhaps the biggest impact is the way retailers can harness the power of social media influencers. Consumers rely on Instagram, Pinterest and Facebook influencers for tips and trends. Syte creates a direct connection between key influencers, the store and consumers.
For example, consumers can use visual search to locate clothes similar to those worn by their heroes and heroines. Or Syte offers an AI stylist platform that discovers products in Instagram images of social media influencers posing in a particular style. Seeing how a fashion icon looks in a slate-grey Astrakhan coat may convince the consumer to press the buy button.
The impact of Syte on revenues is measurable. An ecommerce executive at Boohoo.com, which enthusiastically embraces Syte’s visual search, says: “We see increases in engagement compared to visitors who don’t use search. Conversion rates are 85 per cent higher and pages per session more than 125 per cent higher. Visitors who engage with a ‘view similar’ on the product page have a conversion rate over 100 per cent higher. And the average order value is 12 per cent higher. Lastly, pages per session are 135 per cent higher.”
Naturally, Syte isn’t the only provider of visual search AI. But according to Gartner, it is an outstanding performer, named a Cool Vendor in retail in 2018.
The biggest proof is Syte being chosen by Samsung to power visual search on its smartphone AI, called Bixby.
“Consumers just have to open their camera and point it at something they like. Syte immediately provides them with similar looking items and delivers quality, organic traffic to retailers,” says Ms Pinto Fryman. Retailers can showcase their products on Bixby by signing up to the database via Syte and have access to millions of gen Z and millennial shoppers.
Visual search is a technology retailers must embrace. It removes language barriers, perfect for attracting international buyers. It eliminates the need for customers to know technical fashion terms. How many men know the difference between a pea coat and duster jacket? Visual search means they don’t need to.
Major retailers are all moving towards visual search. Amazon adopted visual search in 2014, followed by Target, Macy’s, Pinterest and eBay. Visual search delivers 48 per cent more product views, according to research by BloomReach. Customers are 57 per cent more likely to return to a site with visual search.
Adding Syte to a retail website is simple. It takes as little as one line of code, depending on the personalisation of the user experience and interface. A software-as-a-service business model, based on traffic and expected volume of users, means the technology is available to retailers and app-makers alike.
“I believe this is the new experience. I see something I like, I simply point at it and I get it,” Ms Pinto Fryman sums up. Visual search is set to change retail, forever. Syte is helping retailers take full advantage of this revolutionary new concept.
To find out more please visit Syte.ai