eCommerce: Text-based search is the problem. Visual Search is the solution.

Marketplaces, retailers, and brands are already aware that most consumers today go online to jumpstart their product search. According to Retail Dive, 87% of shoppers start their product search on digital channels.

Besides finding the products they intend to buy, consumers also search online for brand new products each month. In a Salesforce and Publicis.Sapient study of 500 million shoppers and 1.4 billion ecommerce visits worldwide, “75% of consumers’ site search queries are brand new each month, showing just how fast consumers are discovering new product types, brands, and features.”

As a point of product discovery, search is an important step that can make or break consumers’ shopping experience. In fact, Mike Lowndes, Senior Director Analyst at Gartner shares, “Search is a critical part of a commerce funnel, and can be seen as a low-hanging fruit when considering value-for-money innovation with a potentially high impact on conversion.” (Market Guide for Digital Commerce Search, February 2019)

However, the current text-based product queries are becoming dated to address present consumer demands for convenience, relevance, and choices. Baymard Institute conducted a usability study of e-Commerce search and the results revealed “surprisingly dismal support for essential e-commerce search query types.” Here are some of the stunning results:

  • 70% require users to search by the exact same product type jargon the site uses
  • 22% of the sites don’t support search queries for a color variation (despite the product searched for being available in multiple colors)
  • 60% don’t support thematic search queries such as “spring jacket” or “office chair”
  • 84% don’t handle queries with subjective qualifiers such as “cheap” or “high quality”

The limitations of text-based search are harrowing not only for consumers but also retailers. It is highly dependent on using the right combination of words. When consumers don’t get the correct keywords, or worse, don’t know how to describe in words what they are looking for, it immediately precedes a long and winding product search and overall shopping experience. Retailers, then, miss opportunities to directly connect with consumers and increase revenue.

Visual search is the solution. With this powerful technology, consumers don’t need words to translate what they have in mind. Using only a screenshot or a photo, visual search immediately shows visually similar shoppable products, enabling customers to compare prices, see delivery options, and make the purchase with just a couple of taps and clicks.

According to Gartner’s February 2019 Market Guide for Digital Commerce Search, visual product search is among the AI use cases for digital commerce by top three motivations (figure 2), with 30% of respondents choosing it. Lowndes explains:

Visual search— Search based on an uploaded or linked image. This requires feature detection and pattern matching, so is emerging with the maturity of machine learning algorithms. For digital commerce channels, visual search is a distinct technology from text-based search, and very few vendors overlap. Face recognition and object character recognition (OCR) have been powered by this approach for some time.”

Lowndes continues, “In the digital commerce space, it is especially powerful in the apparel/fashion vertical where ‘shopping the look’ is becoming a popular method of searching.” has also been designated a Cool Vendor in the May 2018 Cool Vendors in AI for Retail report.*

Text-based search is limiting, while visual-based search is liberating. It is the ultimate end-to-end, customer-facing product discovery experience that shoppers are waiting for. Using Syte’s visual AI for retail, marketplaces, retailers, and brands improve consumers’ product search experience, smoothen their whole shopping journey, and eventually guide them through conversion. With happy customers, in return, retailers get high-quality traffic, increase product exposure, and boost basket size and sales.

* Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, express or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.