We use cookies to provide the best site experience.
Ok, don't show again
Close

AI solutions for e-commerce. Visual search

AI-powered visual search is the cutting edge way for e-commerce brands to improve customer experience. No need for exhausting text searches and choosing of suitable wording anymore. Now we can do the search in more natural for human brain way.

It's known, that visual search solves several problems at once, both user and business. Especially in fields, where technical language or years of experience are needed to explain what you need (repair parts, supplies, tools, etc.)

Here are some numbers for better illustration of why visual search leads the pack:

  • 93% of people make a purchase decision based on visual perception of the product;

  • 73% necessarily view product photos before buying;

  • 19% are returned for 19% of search queries on Google;

  • 62% of millennials want visual search over any other new technology;

  • By 2021 early adopter brands that redesign their websites to support visual and voice search will increase digital commerce revenue by 30%.


Additional advantages of using visual search in e-commerce


No need to search for words. How often we can't remember the exact word, brand name, even color! Now you don't need to remember them, because the picture will show you all you need. It means, no language barriers anymore, no misunderstandings – the language of images is understood by everyone, no matter what language the seller and the buyer speak and in which countries they are located.

It's easier for the user and product to find each other. A lot of people want to buy something under the rubric: I saw it from a passerby, friend, neighbor – I liked it – I want it. Without even knowing the name and brand. Far not everyone will want to approach the stranger and ask: "How it is called/Where did you buy it?" Now it's not necessary –point the smartphone's camera, and it becomes immediately clear what the product is and where to buy it.

Neural networks and optical pattern recognition can serve more customers without expanding staff with AI solutions. To prove visual search tactile benefit we used it in our optical recognition system based on neural networks.


Optical recognition for remote controller selling by Evergreen

Sometimes it is difficult for clients to describe an item they need to buy. The client does not know how parts of a product are called, doesn't know professional lexis, but really need to buy the item. It gets even more difficult, when all the items that are sold in your online store, are similar to each other – as is the case with remote controllers.

To make the purchase easier, we have developed an optical recognition system that allows customers to determine the model of the remote control by photo. The earlier business owner needed an experienced manager to choose the model correctly, according to the client's description. But now the chatbot successfully copes with the tasks. The client only needs to upload a photo into the system, and the chatbot recognizes the model and offers a link to the needed remote controller.

More about how we trained a neural network you can read here. To learn about other AI-powered technologies for e-commerce follow the links chatbots, pass recognition and person identification, CUI and personalization.
Made on
Tilda