AI is everywhere and there are & will be artificial intelligence applications for the fashion industry.
We are only scratching the surface with AI and there are various areas within fashion, where AI will and is changing it.
In this post, I want to cover 14 Areas within the Fashion & Retail industry that will be affected by Artificial Intelligence.
1, Supply Chain & Demand Projection
- H&M announced earlier this year that they were sitting on $4.3B of inventory, that is a lot of money that is sitting and being wasted.
- Burberry destroyed $38 Million worth of goods.
- Excess inventory constituted $472 billion of total lost retail revenue globally in 2015, an increase from $362 billion in 2012.
- There is so much money being wasted that could have been prevented if the inventory was managed in a better way.
- Using Artificial Intelligence, H&M & Burberry and other retailers can use prediction analysis that which type of clothes, might not sell and create a combination of discounts, free shipping, product combo’s etc. to clear out the inventory that is most likely not to sell after a certain time period.
- Rue La La collaborated with MIT to experiment with AI technologies to improve demand forecasting.
- They developed a system that helped them predict the demand for products in their flash sales and accounted for data sources including brand information, product type, color, price and a range of other factors.
- This enabled them to optimize prices and generate a 10 percent increase in revenue without the extra burden of unused inventory or supplier costs.
- German e-commerce player Otto uses deep learning to analyze billions of transactions and is 90 percent accurate in forecasting what it will sell in the next 30 days.
- This insight allows it to order a couple of hundred thousand items each month from vendors with no human intervention. It has also cut surplus stock by 20 percent and reduced costly returns by 2 million items each year.
2, Photos for a new collection
- When you create a new clothing line or launch a new collection, you have to take photos of all the items to be posted on your website.
- It can be extremely cost sensitive because it involves models, photographers, makeup artists etc and you have to manage their schedules and re-shoots can be a pain in the ass.
- There are AI companies that are enabling retailers to shoot images by putting clothes on mannequins and then replacing the mannequin with the 3D Image of the actual model. eg:
- This way, retailers, fashion designers do not have to pay for expensive photoshoots
3, GAP: Sorting Orders at Warehouse
- When you order a product on GAP or Amazon or any other online retailer, there are actual people on the other side picking and sorting products for you.
- It is a huge cost for because retailers have to pay them, it takes a longer time to sort products and most warehouses can’t function 24/7.
- GAP is testing and working with Kindred AI to train robots to pick and sort its products.
- Imagine: Robots pick items of different sizes, puts them together in a nice box and drone brings it to your door.
4, Inventory Checks and Re-stocking
- When you have a clothing store, how do you find out you need to restock the items on shelves?
- Today human beings go visit the racks and check if items need to be restocked.
- Then, they go to the back, pick up items, bring them to the front and restock them. This is very labor intensive and costs money.
- Camera’s and sensors can automagically sense what is missing and need to be restocked and can order it for you.
- And they can pair the understanding of the seasons, sales data etc to make sure they only order the item that is more likely to sell.
- And instead of getting new items manufactured, they can check if that item or similar item exists at other locations but its probability of selling at that location is low it, it can order it from there or ship it to the customer from there.
- Robots can be used to pick up clothes from the back and re-stock on the shelves. This will save lots of labor cost and employees can focus on helping customers vs doing mundane tasks.
5, Artificial Intelligence Fashion Designer
Amazon, Stitch Fix, and Myntra have developed AI that now designs clothes for them to fill the gap.
- Amazon is developing AI that combines hundreds of thousands of designs, fabrics etc to create something new that it interprets that could sell.
- Myntra’s brand Moda Rapido is powered by AI and works without human intervention to offer computer-generated designs, including T-shirts, jeans, kurtas, and shoes. The system is fed data from various sources, including customer data, social media, fashion publications etc., and creates thousands of combinations of designs, then hones in on what would sell well. It now has the highest gross margins compared to all other 14 brands under the Myntra portfolio.
- Stitch Fix:
- Stitch Fix uses 3 set of Algorithms to create designs via Artificial Intelligence & Machine Learning.
- According to Eric Colson, Chief Algorithm Officer at Stitch Fix
The team identifies viable gaps in the company’s inventory – clothes that people would buy but a designer hasn’t made yet.
You can find a detailed post on How Amazon, Stitch Fix, and Myntra are using AI to replace Fashion Designers.
6, Trend Forecasting
- Trend forecasting is big business. There are companies such as WGSN does that. They have employees that go thru thousands of images, look for patterns, new trends and sort them out to do trend forecast.
- The problem is that not every designer can afford to access WGSN’s data because it costs more than $40K per year.
- AI can pick images from Top bloggers, Instagram, Pinterest etc & analyze which images, colors, style of photos that didn’t exist before but are getting high engagement.
- And then combine that with data from hashtags, location, influencer, color distribution etc.
- Artificial Intelligence can help analyze & predict upcoming trends.
7, Dynamic Retail Website
- Each one of us is unique and often is influenced by different words, colors, images, models, website layouts etc.
- AI that can dynamically alter the product images, product descriptions based on which user is browsing the website.
- The AI company Sentient Technologies works with Sketchers to improve recommendations based on analyzing which images of shoes an individual customer prefers to curate a gallery tailored to a shopper’s unique style preferences
8, Product Testing (reduce returns and wrong purchases)
Getting product in the hands of users so they can test it out before they purchase, can get expensive.
- This about Warby Parker, when you order prescription glasses, they send you 5 frames that you can test it out at home and then you place your order and pick one pair that you like the most.
- Which means, Warby Parker has to keep a separate set of inventory that it sends to its users for testing and pay shipping cost both ways.
- Take your image and you can apply Prescription glasses filter on your face. It will save these companies ton’s of money and easier for consumers.
- Sephora’s makeup app: Apply makeup, liners etc on your face via this app and you won’t buy the wrong product.
- Hair Style: MODFace, enables you to a try different type of hair colors. Next step could be to try a different type of hairstyles to see how it would look on you.
(image courtsey: MODface)
9, Clothing recommendations & personalization
- Stitch Fix is a prime example, where its algorithms analyze returns & buy from its customers and then sending them clothes that are more likely to buy.
- This data is also important for brands because before they didn’t’ know because this type of activity happened in fitting rooms.
- There are other companies that use AI to recommend products to customers eg: LeTote, BirchBox, Rent The Runway, IPSY etc.
10, Virtual Fitting Rooms
(image courtesy: bevond.com)
- This is probably the most obvious application of artificial intelligence in the retail industry.
- Uniqlo, Burberry, Memomi, Amazon etc have been working on creating a seamless virtual mirror so consumers can try new clothing, makeup items virtually.
- Vision Technology can be used to create virtual fitting rooms enabling customers to virtually test out various clothing items.
11, Eliminate Shoplifting
(image courtesy: thecollegefix)
- Amazon Go created a store where you can walk in, grab items and then walk out.
- Companies such as AiFi are creating a new type of seamless shopping experience. Where you walk into the store and the cameras recognize whom you are. As you pick any item it records you as the owner and adds to your list. When you put the item back, its removed from your list.
- When you walk out of the store, with any number of items, they are automatically added to your credit card.
- The system keeps track of which items you picked and put back. This way, if people try to shoplift products, and sneak out, they will automatically be charged for it. Therefore, Artificial Intelligence technology can eliminate shoplifting.
- Amazon Alexa & Google Voice are taking the lead in the voice-based ordering system.
- You can talk to Alexa and have it order any items you need from Amazon.com
- It is only a matter of time when the users would be able to view items on their TV and order directly from TV. Because Google Home, knows where you live, it has your credit card, as you use it more, it will know what type of products you like and learn about your shopping habits.
13, Out of Stock
- Imagine you have an online retail store and your items are so popular that you are “out of stock.” A potential customer visits your site and sees’s “out of stock” and they turn away.
- Using AI based recommendation engine, you can give that customer, best possible alternative for the item that is out of stock.
- Picalike is another example; it collaborated with BonPrix to make personalized recommendations for out-of-stock items based on visual similarities, reducing the dropout rate on out-of-stock items by 43 percent.
14, Clothing Insurance
This is one of my favorite artificial intelligence application for the fashion industry that is not obvious.
- Fashion designers, retailers etc buy insurance from larger companies like Allstate, Nationwide etc. And its usually calculated by national average.
- Imagine an AI solution that creates specific insurance just for you based on:
- who your customer is,
- how many returns you get,
- how many collections per year, or
- if you decide to skip a collection or reduce the number of returns then it can reduce your insurance
- This would or could enable Personalized Insurance based on your activity and could help you save money.
As you can see, there are multiple areas where Artificial Intelligence applications could affect the Fashion industry.
And this is just the beginning.
As the technology gets more advanced, we will see lot more areas where consumers will start seeing AI.
I would love to hear from YOU about which areas you think artificial intelligence could affect the retail industry.
Watch Video: 14 Areas where AI is changing the fashion industry