Retail industry has one of the most extensive value chains. It spans across design, sourcing, manufacturing, distribution and fulfilment and needs constant, steady innovation in each of the channels. No wonder it is one of the largest industries in India at almost 1 trillion USD. But, even today, 85% of retail happens through unorganized channels which creates a huge opportunity to optimize the value chain using artificial intelligence.
One of the most complex verticals of retail is fashion. Is it art, is it science or is it purely a fad? Since long, fashion has been dictated by a few, influential people in the industry and all others just follow suit. But, this has been changing slowly. Businesses in fashion retail are leveraging data at every step, be it trying to cut losses, increase sales or retain customers. Leveraging data means listening to the consumer, in one way or the other, and predicting what they’ll want to buy in the next season. In this article, we cover the various ways in which data and artificial intelligence is being utilized to solve some of the biggest challenges faced by the fashion industry.
1. Design: While fashion retail is a behemoth in itself, it comes with a lot of competition. Brands have to remain on their toes just to stay afloat in the market. This means coming out with the latest trends and designs in a very short span of time. Fast fashion retailers look at the runways at fashion shows and make clothes really quickly in a “see now-buy now” retail environment. This involves a lot of number crunching to do to analyze trends and customer behaviour and predict the sales output of a particular design. Companies like Zara and H&M have nailed this type of business model and can bring in new designs to market with a lead time of 4–6 weeks where the industry average is around 36 weeks.
2. Inventory Management: Fast fashion not only requires designs to be brought to the market in an extremely short timeline, but also makes inventory management difficult. Some would claim that inventory management was a problem in the previous century, but even today 20% of the clothes produced never get sold. They end up in incinerators where they are burnt, making fashion one of the biggest polluters in the world. H&M alone burnt USD 4.3 billion worth of clothes in 2017.
3. Pricing: This is a function of other factors such as demand prediction, cost of manufacturing and competitor’s pricing. Companies run machine learning algorithms on a combination of these factors to find the ideal price points at which they can sell various products while maximising revenue. This is why you can see a stark difference in pricing across platforms like Flipkart, Amazon, Myntra. You can find the exact same dress on these platforms with a price difference of almost INR 600–800.
4. Cataloguing: Ever wondered how product related information like color, design, sleeve length, collar, etc. are tagged to thousands of products online? If you are thinking of a team of 5 people sitting behind computers in a warehouse, you are 5 years into the past. These days companies leverage image recognition techniques to label various attributes of a product. One of the companies that pioneered the automation of cataloguing in fashion is Playment, which crowdsources information on various products and then uses that as a data input to run their algorithms.
5. Customization: This seems like a no brainer, but e-retail companies have been working hard to make their recommendation tools provide an extremely customized experience to their customers. The customization problem is not as simple as it seems. Algorithms track customer journeys to help them find the right products. But, there are multiple parameters that go into a recommendation tool and more often than not, these endless parameters generate recommendations that might not be that relevant to the customer. Companies are using data from multiple sources to add this layer of customization. Unbxd is one of the leaders in providing recommendation tools to online retail businesses that do not have access to the level of data and technology needed for customization.
6. In-store experience: AR/VR is not just for the gaming industry. Fashion has started adopting AR/VR to improve customer experience significantly in brick and mortar stores. It helps customers visualize what a product would look like on them without actually having to try it out. Say goodbye to long queues for the trial rooms. Luxury brands like Gucci, Burberry and Vogue utilize Apple’s AR kit to provide such experiences to their shoppers. This technology has started seeping in online fashion as well with multiple platforms running pilots for the same.
7. Customer Service: A significant chunk of post-sale customer service is automated using chatbots which utilize basic principles of artificial intelligence and natural language processing. Fashion retailers have turned to technology to recreate the personalized atmosphere that one would find at a brick and mortar store. Tommy Hilfiger’s chatbot makes it seem like you are actually talking to a sales executive at the store. There are multiple companies operating in this domain who provide easy to integrate chatbots for retail.
While online fashion retail is still in its nascent stages in India, Gartner predicts that by 2020, customers will manage 85% of their relationships with enterprises without interacting with a human. There is still a lot of room for improvement in terms of customer experience within online shopping. Parati is a product discovery app which aggregates products from across platforms and allows users to shop from a personalized feed of clothes. It is a customer preference driven app and generates recommendations for users that fit them and their personality. You can check out the website www.parati.in for more information and download the app from playstore.
