(This is a guest post from our friends over at Impresee, a visual API company that offers services for eCommerce, such as visual search and creative search.)
For some time now eCommerce has been around and growing, and it isn’t showing any signs of slowing either. In 2014 online stores worldwide sold around 1,300 billion US dollars and they’re expected to sell 4,800 billion US dollars in 2021!
With figures like those it’s clear that nowadays the problem eCommerce faces has nothing to do with getting people to buy online but with how much online competition has grown! Not just smaller businesses competing against each other, but industry giants squeezing out smaller online stores too.
As the owner of a small online store, there’s no way you can compete with the giants, and you shouldn’t try! But how do you manage not to compete? Well, the answer is simply customer engagement.
Customer engagement is having your customers return to your business every time for each new purchase. But how do you do this? How do you engage your customers? From my point of view there are two ways in which you have to work your site to engage customers: make it useful and make it fun.
Making a site useful is all about making it easy for the clients to get what they want without getting frustrated, so having a good search engine, utilizing a good chatbot, among others. In other words, it’s about efficiency and effectiveness.
Making it fun has to do with how much time customers will spend navigating after finding what they wanted in the first place.
In this article, I’ll give you some ideas on how to use AI to make your site more useful and more fun to visit.
Make It Useful
As I mentioned before having a useful site has to do with creating a good experience for your customers, by allowing them to get what they want in an easy and fast way. In this sense, AI has a lot offer but today I’ll show you only three examples, these are: Visual search, recommendations and improved chatbots.
The idea behind visual search is to give customers the chance to show the online store what they want, which translates into uploading a picture and searching by it. This means that instead of describing what they want with words, your clients; will show you what they want, and, in exchange, you’ll give them a list of the most similar products you have in stock.
Recommendations have been around for a while. They generally work in two ways: recommendations based on the things you’ve previously bought, and recommendations based on what other people have bought. An example of how AI can be used in recommendations is to determine a relationship between the products in your online store. For instance, let’s say someone buys a casual green t-shirt. Here, by building an AI model you could automatically determine that you should recommend a pair of blue jeans to that person. This means; that AI can give you new ways in which to recommend your products.
Anyone who has had a problem in an online store knows how frustrating it can be to speak to some chatbots, because even when they should appear to be human beings they don’t respond as such. Luckily for us AI can help us build “smarter” chatbots, which would actually be able to interpret our words without the need of using specific keywords. And not only that, AI also gives the chance to have multimedia chatbots, that is bots that don’t only interpret text but can also understand the images a customer uploads, or even audio.
Make It Fun
We all know that the idea of an online store is to keep people interested, and this means having them look around as if they were in a brick-and-mortar store. While there’s no magical solution that’ll keep any kind of customer interested, I can tell you that making the experience inside your site fun will most certainly boost customer retention and return visits.
I’ll now present you two ways in which AI can help you make your site fun, these are: by using augmented reality (AR), and search by design.
The IKEA catalog app allows customers to use AR to locate a piece of furniture inside the living room, so that they can see how it’ll look. But, what if your customers could tell the app what they’re looking for and have it place the resulting item in t
he room? For example, a customer could request a sofa option, and the app would place a sofa that matches the color of the floor, or the style of the curtains of the room. That would be mixing AR with AI, and it’s fun for your customers.
Search by design allows your customers to look for products by using their creativity. It allows them to draw and paint what they want, using that within the online store setting, which then presents them with the most similar products in stock. Of c
ourse, search by design is useful when customers want products that are difficult to describe, and which they don’t have photos of. Nevertheless, what’s more important is how people use it and the emotions it generates, especially when they enjoy the satisfying feeling of doing a rough sketch and discovering that the system recognized it well enough to actually find them the products they were searching for. That’s fun!
The truth is that, for eCommerce at least, the amount of uses one can find for AI is almost limitless. However, the most important point is not to include as many as you can in your eCommerce but to include those that match your customers and, of course, your business.
About the Author
Camila Alvarez is a Co-founder and computer vision researcher at Impresee, a visual API company that offers visual services to eCommerce. Camila focuses on finding new solutions for the fashion and apparel markets to increase customer engagement in these industries.