News & Views / AI Goes Shopping

by Contagious I/O

This story originally appeared on Contagious I/O, our customisable research platform featuring the world’s most innovative, creative and effective campaigns and marketing ideas

Canadian shoe site uses artificial intelligence to power visual search

Online retailer is using artificial intelligence to help shoppers find their perfect pair of boots.

The Vancouver-based ecommerce brand has partnered with AI experts Sentient Technologies to create the Visual Filter. The service launched on its Canadian site, this week.

Instead of typing into a search bar or checking filter boxes, the tech learns what the customer likes by analysing the products they click on. Selecting a pair of heeled boots, for instance, brings up more heeled boots. Judging by the images that attract the customer, the retailer can then determine that she might be interested in black rather than brown shoes, or shoes with laces rather than ones without, endlessly bringing up more styles that match their preferences.

The product images don’t contain any words at all, until shoppers click for more details where they find out about available sizes, pricing and brand. They can then choose to save or purchase the product. describes the process as being similar to working with a personal shopper, but the system relies on a visual communication not a verbal conversation. is currently testing its Visual Filter with women’s boots but has said that it will expand the service across its platform in the near future.

Sentient Technologies has said that the technology could be adapted to products beyond shoes and clothing, including furniture, housewares and cars. The company, which was founded in 2008, has received $143m in funding. Its founder Antoine Blondeau has proven experience in the area of artificial intelligence as he helped develop the tech behind Apple’s virtual assistant Siri.


Contagious Insight / 

More than words / According to, there are a few key reasons why it’s difficult for shoppers to find exactly what they are looking for online. Firstly, people can’t always express exactly what they’re looking for in words, making it tricky to type a description in a search bar. Secondly, it’s hard for retailers themselves to accurately translate photos of their product offering into words. And finally, often people don’t know what they want until they see it.

As Nikki Laing, buyer said: ‘Many consumers don’t know the best way to describe the shoes they want but they know them when they see them. Visual Filter allows our consumers to enjoy a highly interactive discovery process that’s as simple as pointing and clicking. The results are so insightful, it really feels like magic.’

In its explanatory video describes its tech as like shopping with your best friend: it knows what you like and what you don’t like, without having to explain why. And by using artificial intelligence, the retailer provides a unique experience that you couldn’t recreate by simply searching on Google.

Too much choice / Explaining why its tool is so useful for retailers, Randy Dean, Sentient head of corporate and strategic development cites the paradox of choice. As he explains in the video above, merchandisers need to offer a large selection of products to appeal to a wide audience, but this also means that it’s difficult for shoppers to find exactly what they’re after.

Smarter discovery / Sentient’s software allows for a very personalised discovery process. This works is sharp contrast to other algorithmic tools to promote discovery. The ‘customers who bought this item also bought’ tab, which is used by sites like Amazon, helps narrow the paralyzing choice of products to a small pool of items. But it also artificially promotes by redirecting everyone there, even when the shopper might be happier with an entirely different item.

Visual Filters helps shoppers navigate the endless aisle, i.e. a brand’s entire catalogue, with purpose. Shoppers can at once scroll endlessly, as they would using a site like Pinterest, but their search results becoming more accurate with time.’s co-founder and chairman Roger Hardy, suggests that the Visual Filter tool also encourages people to buy from brands that they are unfamiliar with. ‘What makes this attractive is that people can get to the shoe they love without knowing what brand it is,’ he told The New York Times. ‘If I told you there was an Italian company with the perfect heel, toe and lacing for you, but didn’t know the brand, it wouldn’t do you any good.’

Visual search to shop /’s launch of its Visual Filter features comes as more companies are experimenting with ways of improving visual search to better the shopping experience. Pinterest, for instance, has just launched an image recognition tool that lets you use a photograph as a starting point for a search in order to bring up similar items that you can buy. Apps like Donde Fashion let people shop for clothes online by relying on visual search rather than text. And others like Snap Fashion and Slyce let people take photos of clothes they covet to bring up lookalike products. As visual search to shop becomes a trend, it’ll be interesting to see how much better retailers get at figuring out what we want to buy without shoppers using any words at all.



This story originally appeared on Contagious I/O. Contagious I/O is our bespoke trends, inspiration, insight and analysis service, providing daily innovative marketing intelligence across a comprehensive range of sectors to brands and agencies across the world. For more information about Contagious I/O contact