Visual Search is Revolutionising the way we Shop

Over the last few years, visual content has slowly been replacing text to reach its current position as the most important online content. Social networks have been leading the way with this concept shift; in the last year, Instagram’s active user base grew by 64% and that of Pinterest by an incredible 111%.

While we’ve spent the last few years getting used to this richness in visuals and imagery, it’s also now affecting the way we think and interact. Unsurprisingly, many eCommerce companies have already taken measures to place imagery as a number one priority. Our own consumer research revealed that 52% of shoppers say product images are the most important aspect of an online experience.

It’s perhaps unsurprising then that this notion of visual content supremacy is naturally transitioning across into eCommerce and shoppers’ minds and with it a desire for a new search trend focussed on image . One that not only enhances but also simplifies the search experience . The once incredible possibility of buying a desired product using just an image has become a reality. Take a photo of a skirt you like in a magazine or on the street and that product can be on its way to you within minutes.

There are many benefits of visual search within eCommerce, for both consumers and retailers, and here I’m going to go through a few.

1.Connect customers minds more directly with the eCommerce catalogue

Letting customers search using image recognition allows them to find any product independently of the source of inspiration in which the item was discovered; be it a printed magazine, street style trends or a social network. It means that either the exact product or very similar products will be shown, and the findability rate, the ranking of the most relevant products on the results page, will increase.

Moreover, it’s not always easy for customers to describe with words the specific product they have in mind. What’s more, different shoppers will most likely describe an item in a different way and then each eCommerce site will also use their own terminology to describe a specific product. By incorporating visual search technology into the search feature, the connection between the customer need and the product catalogue is not only easier and more straightforward but also extremely more effective and accurate.

Take the following example. Emma is looking for a maternity dress she saw on Instagram. She knew the brand, so she typed “maternity dress” on the online shop and while she found lot of results, none corresponded with the dress she had seen. She tried again, drilling down the query with more detail and typed “dotted maternity dress”. Again, lot of results appeared but not the dress she was looking for, which was incredibly frustrating as she knew it was within that brand catalogue.

This experience could be infinitely improved if all Emma needed to do was take a photo of her inspiration dress from Instagram and use a visual searching tool… and boom she finds the exact dress she was looking for! In this instance, the issue was that it wasn’t in fact a maternity dress, but a dress that fits very well on a pregnant woman like herself. With visual search functionality in place, available from the search box, the query is much more precise and results will correspond more accurately to what is being searched.

In the same way, image recognition is also a useful tool as part of an omnichannel strategy , supporting the physical store in a similar way that a customer might search with a bar code or reference number on their phone to find stock availability online.

The quickest way to purchase an item that’s unavailable in store is, rather than typing a description or having a guess at what you think the product may be called, to instead take a photo of the item and immediately check its online availability in the size or colour you want. To enhance the experience, and another benefit of visual search, is that the online store could go one step further to optimize the results offered by providing not only the searched product but also similar options or matching styles.

2. Find similar products or be inspired

Let’s not forget that search isn’t just about finding products, a key, and ever-more important, element of search is that of discovery. Shoppers increasingly want to be inspired and receive recommendations or style ideas. A third (33%) of the consumers we questioned for our recent survey said this was something they wanted every time or a lot of the time when they visited an eCommerce store. Visual search opens up a whole new world of possibilities here.

I still remember when fashion bloggers were dedicated to looking for similar products or even clones of those that were either unaffordable or sold out. Through the use of cutting-edge visual recognition technology, we can all find similar products or those within the same category, style or colour range very simply. This becomes even easier on low-cost retail sites that are dedicated to the early detection of trends and who translate runway collections to the high street to offer them to the general public.

ASOS offers similar products when using visual search

Additionally, the retailer themselves can offer recommendations , or a ‘shop the look’ type service, based on visual matches to amplify and enhance the product discovery experience. This means shoppers no longer need to go through categories or type new queries, especially if they don’t realise that product matching exists. Colour-matching is also an attractive functionality of image recognition as it means recommendations can be automated by filtering through a broader palette of colour shades.

3. Tagging and content curation

Visual search is not just about potential customers using their mobiles to capture images and scan products. Image recognition is also becoming a faithful ally for retailers. It enables every product to be automatically identified, categorized, classified and tagged in the backend. With each item given attribute labels for their physical and colour qualities. This is extremely useful for retailers as it’s much more efficient, saving on time, labour and costs while improving and speeding up the catalogue management.

What’s more, it also improves the relevancy of the results by classifying each image according to its predominance of visual aspects such as style, colour range or even the occasion the product is suitable for. Companies like Pinterest and Amazon were early adopters of this new way to search and shop, and it’s not surprising that big fashion marketplaces like Zalando or Asos have quickly followed suite to incorporate visual search. But this is just the start. Technical fine tuning on image recognition capabilities and functionality is swiftly progressing so it’s now only a matter of time for visual search to be widely embraced. The uptake and popularity of this technology is set to be one of the big eCommerce trends of 2019.