Artificial intelligence makes finding YOUR style online fun and easy
Octubre 22, 2018
Author: Camila Álvarez
I'm a co-founder at Impresee. I like researching deep learning techniques and their applications in the style and apparel industries. You can contact me at camila.alvarez@impresee.com

How artificial intelligence is related to fashion industry?
Style is all about self-awareness: what kind of clothes look good on me? For what occasion? Which colors should I wear?

Once you’ve figured that out, there’s no need to even leave the house! Because with artificial intelligence, you won’t even need to try on the clothes! This is just one of the reasons why the style and fashion industry has become such a big player in the online market. But, although e-commerce has its advantages when compared with physical stores, there was always one big problem – it can be hard to find what you want.

We’ve all been there – you go to buy something, let’s say a red dress, on an online store. You put your search term in the box, and use the categories menu to define the clothing type… but, although you get a huge number of results, none of those red dresses are exactly what you need. How frustrating! Now imagine if you could somehow tell the system you’re looking for a red dress, with a high collar in the back, with an open sweetheart neckline. This is what artificial intelligence can do for your customer.

So, now you’re asking yourself: how can these online stores give their clients an experience so good that they’ll keep coming back for more? How can they help their clients find what they want?

We’ll show you how AI can resolve these questions. In particular, we’ll introduce three ways in which to improve customer engagement and experience, which in turn will produce continuous market growth and extension.

Artificial intelligence & Visual Search

Put on your style-consumer outfit and imagine you’ve just seen a pair of shoes you really like, you even have a picture of them. In this case, one of the main advantages of visiting a brick-and-mortar store is that you can get someone to help find you want, you can describe what you want or even show the store assistant the photo or picture you have. That gives them a good idea of what you’re looking for and allows them to think of something similar that the store may have in stock. You could also try using your photo in a standard search engine, along with keywords…  But you know it’s not going to be effective.

Visual search provides that sense of human communication but skips the part that makes it more difficult for the customer: finding the right words to describe a complex desire.

By including a visual search element in the site, the online store gives its customers the opportunity to communicate with the site via images, and the site will respond by showing them the products that are most similar to what the client wants. Fast, effective and simple!

Creative search

Your customer has spent a whole week thinking about a dress. She needs it to be perfect and hasn’t seen anything similar in physical stores, and of course, she doesn’t have a picture of it. She has it in her mind and she has to find it! What does she do? If she’s got the time and energy, she could go to a store and get someone to help. If she’s lucky she might even find what she’s looking for, probably after visiting several stores, but if she’s unlucky… well, let’s not talk about that.

But, what if she can’t go to a store? What if she doesn’t want to go? In that case, she’ll probably start by looking online. She’ll start searching by giving a text description of that perfect dress. Soon enough she realizes it’s not working. She’ll end up frustrated and won’t be any closer to having found that dress.

How would this change if one of those online stores allows that customer to design that dress on their site? It’d be like showing a designer sketch to a person in the store, right? The idea of creative search is to give the customers the possibility to visually communicate with the site not just with photos, but with drawings as well. This way the customers are no longer required to look for things they have actually seen, but for items, they imagine and truly want!

Not only do the customers are benefited, the store wins too because this technology allows them to see what people are really looking for.

Style advice

You have an important meeting next week. You’ve nearly got the whole outfit but you’re missing the shoes. You don’t know which shoes you want, only the occasion for which you need them. You go to your favorite online store and start going through the shoes category. Page after page after page. There are more than 50 of them but you keep going. You’ve reached page 38 and you feel like you’ve spent your whole life searching! You only want shoes that go well with that outfit and that are appropriate for the occasion! AI can help you with this.

Style advice tools allow the customer to communicate to the store what they’re looking for, as well as the occasion and clothes with which the missing item must match. Using this information the tool will search for items inside the catalog of the online store and will show the customer only those that match the criteria. So, you give the store your information and voilá! a page appears with a selection of footwear, perfectly matched for the occasion and the color and style of your outfit. It’s like having your own personal style assistant – without the cost, without the stress, and without having to leave the comfort of your own home. Job well done.