Have you ever had a moment where you wanted to buy something you saw but did not know how to put it in words to search for it online? Visual Search is here to help.

In simple words, Visual Search helps users to search for any product using its images. So, instead of typing “Red Crepe Gown with side-slits”, now you can just upload its picture and get to the identical product you’re looking for with bunch of similar product suggestions.

Visual search in ecommerce was first used by Zappos in 2009. And then again 2010, visual search became a news when Like.com — the visual search engine that helped users find similar products on the web and compare their prices was acquired by the search engine giant Google.

Why Visual Search over text search?

With the existing text based search, keywords must be there in both the query and the product description or the metadata on the page. The match depends on product content that describes the product in the same way. Visual search eliminates these dependencies and does an image to image comparison to get identical as well as similar results. Due to this, visual search will always have a conspicuous advantage over keyword-based search in the eCommerce space.

Where is Visual Search most helpful?

Whatever you buy because of its aesthetics & looks and which cannot be put into words is where Visual search can come in handy. This includes broadly following categories:

  • Fashion (Clothing, Jewelry, Bags, Footwear, etc.)
  • Home & Kitchen furnishings (Furniture, Decor, Housekeeping, Kitchen, etc.)
    Together the above account for more than 50% of the worldwide E-commerce revenue.

Visual Search in E-commerce

There are 3 ways in which visual search is powering ecommerce websites today:

  1. Style-Based & Popularity-Agnostic Product Recommendations: Suggest products with similar pattern shape and colour from your entire catalog. Reduce your overall “Cart-drops” and retain your customers by providing visually similar product recommendations.
  2. Click and Search: Give your users ability to click and upload photo to find what they are looking for.
  3. De-Duplication: Identify duplicate products and merge them to clean and optimize the catalog.

How does Visual Search help E-commerce?

1. Increased product & complete catalog discovery

For an eCommerce website, only about 30% of the whole catalog is discovered by customers with the same products being displayed again and again with similar kind of searches. The rest of the catalog remains undiscovered due to lack of good search and recommendations. With Visual search, the whole catalog becomes open for discovery.

E.g. when you search for a brown L-sectional Sofa using text search, you might always get more or less the same results on a particular website. But when you search it using an image, you will see only those results that are visually similar to the searched product.
An eMarketer study with data from Bloomreach reveals that visual search on eCommerce sites leads to 48% more product views, up to 75% more return visits and up to 11% increase in average order value

2. Social media influenced purchase

According to research by Accenture, social media will become the preferred shopping channel for Generation Z, with 69% of young consumers interested in purchasing directly through social networks.
Visual Search will become an indispensable tool for the eCommerce players to tap this young market segment. By having a visual search option, e-com players will enable their customers to upload and search for products that they find on social media websites.

3. Cross-sell

Visual search is also an excellent cross-selling tool that returns visually similar products in case the website lacks the desired product in stock. Visual search also reduces basket abandonment and saves the users from the annoying process of vaguely searching for products

4. Better lead conversion

Visual search reduces the search time for consumers and hence quickly takes him/her towards checkout leading to better lead conversions.

Brands already using Visual Search successfully

Amazon plugged in visual search into its main iOS app in 2014. Another eCommerce giant — Ebay with 1.1 billion listings on its platform rolled out its “Find It On eBay” in Oct’2017.

John Lewis' Find Similar iPad app lets shoppers search for items based on a product’s shape, colour and pattern. It was launched after a trial in which 90% of customers said they found it useful.
Some key brands (both in retail and eCommerce space) that have successfully used visual search in US and Europe include Macy’s, Target , ASOS, JCPenney, ToysRus, Urban Outfitters, West Elm, Zalando, etc.

It’s being predicted that 50% of all search queries by 2020 are likely to be either through images or speech. So more brands and retailers are likely to turn to visual search as an essential eCommerce tool across the globe.

What are your thoughts on Visual search in eCommerce? Write to us at mail@turingiq.com and check out our Visual Search here.
Look out for our upcoming blogs/articles to understand how Visual search is becoming an essential part of customer search behavior.

“Turing Analytics is a machine learning company that delivers intelligent solutions to Retailers helping them improve product discovery, customer engagement and boost conversion rates. They expertise in building Visual Search & recommendation solutions, which enable retailers to recommend visually similar products to their customers and search products by uploading a picture.”