Visual Search for 4 Million Items: How Intu Does Syte

Intu is a UK company that brings a superior shopping experience on and offline. Not only are they the UK’s largest online shopping centre, curating content and offering retail sales for UK consumers online, they also have 14 physical locations.

intu visual search

In short, they sell a lot of stuff. And their huge inventory from hundreds of retailers meant that intu needed a way to quickly get their customers to the items they wanted to shop for online. This led them to Syte’s visual search, which they now have on their website

“It is a new way for consumers to discover products,” says Karen Harris, Managing Director of intu Digital. “We wanted to create better discoverability on our site for our consumers, because we have 500+ retailers integrated with us and over 4million products.”

Screen Shot 2018-05-11 at 2.02.27 PM

And once they started using Syte to help their consumers, they quickly realized the potential for image search to transform their business. “Image recognition technology is moving so fast and has become so good that we are now hoping to use it not only for online visual search for our consumers but also to help us deeply categorize our retailer’s products within our taxonomy,” said Harris.

For intu, an affiliate publisher that integrates with hundreds of retailers and brands, the sky’s the limit for image technology. “Visual search was a huge step forward for us, and I love that you can scan through pages to find a style you like and know that a similar style will appear immediately. Visual technology is a massive advantage for us, and we want to continue using the technology as it develops.”

A Picture’s Worth a Thousand Searches

Intu works with hundreds of retailers to bring a huge range of options to their consumers. But this comes with challenges that all ecommerce companies can relate to — helping customers navigate efficiently to where they want to go. Put succinctly: “Discoverability, when you’ve got over 4 million products, is really hard.”

Image search takes the hassle away from consumers who would otherwise be bogged down trying to figure out what to search for and how to sort options to find exactly what they want. Instead, consumers can just hit the “photo search” button on intu’s website and choose any screenshot or saved picture from their desktop:

Screen Shot 2018-05-11 at 2.25.53 PM

Intu returns items that are similar to those shown in the photo — in this case, the exact dress and several more — and lets consumers choose the price and brand options to narrow down their shopping options:

Screen Shot 2018-05-11 at 2.26.07 PM

Going from hundreds and hundreds of dress options to a handful of curated selections takes a huge frustration out of the customer experience.

But it’s not just the customers who are benefiting from Syte’s technology either.

“Organizing data and tagging items for search is particularly hard when you’re an aggregator, like us, because we don’t control any of our data,” says Harris. “We don’t control the imagery — we receive data feeds from our retailers and often that data isn’t good enough to be able to create any kind of discoverability on our site for consumers.”

“Discovery, when you’ve got over 4 million products, is really hard.”

Manually indexing each item they sell would take a huge team dedicated to tracking down information and adding metadata to millions of items round the clock. But that problem is circumvented by Syte’s look-alike and photo search features.

Instead of being able to accurately tag every item down to color, cut and fabric type, intu can use the data they get to present options and use Syte to direct customers once they’ve found something they like. Say you’re searching for that summer dress. You put in maxi dress and check off blue for color, and as soon as you see something you like, you can immediately be taken to dozens of similar items:

Screen Shot 2018-05-11 at 2.34.52 PM Screen Shot 2018-05-11 at 2.35.04 PM

These dresses may or may not have had the metadata to get them into on-site search, but that doesn’t matter when a customer can easily navigate based on the qualities of the image itself, not just the qualities attached to it.

“Anyone I’ve demoed this to has been amazed by it and said it’s an incredible feature,” said Harris when asked how people were reacting to the visual search features. “You can scan through pages and find a style that you like, and all of a sudden, similar styles are there for you immediately, and Syte works particularly well because of the breadth and depth of product that we have.”

Towards the Visual Future

Image technology has become such a key part of intu’s online business that they already have plans to expand it in the future. They want to have look-alike options for their entire inventory, giving consumers better access to their items regardless of what they search for.

And they also see the potential for different types of consumer-friendly applications to develop, like Syte’s “shop the look” feature, where consumers can see an outfit and immediately shop the items shown, as seen here on Fashion Lover:

Screen Shot 2018-05-11 at 3.11.25 PM

“We want to be able to pull together the accessories, the top, the trousers, the shoes, the earrings — all the elements that stand out — and be able to curate either those items or similar to those items, to create one look.” This type of shopping feature has broad applications across ecommerce, and Intu knows that pushing out their visual shopping tools will only benefit their consumers.

“Visual technology is a massive advantage for us, and we want to continue using the technology as it develops.”

Intu also sees image recognition technology as being able to help their own internal processes down the line. It capitalizes on the most faithful representation of products — images. “In terms of being able to identify elements, visual search was a huge step for us, because the one thing we did have from our retailers was a product image. But if we can then add more meta-elements internally to develop our site and our features, that’s another step that we want to take as the technology develops.”

Building Better Ecommerce, Together

Syte knows that image recognition technology can open up new worlds for ecommerce and push the boundaries of AI capabilities. Developing solutions for our partners and seeing image technology shape the way consumers behave is behind the scenes of everything Syte does. “One of the things I’ve really liked about working with Syte is they’re quick, and nothing is a problem.” These types of relationships are what push the industry ahead, and Syte is excited to know what we can do for you. Contact us today.