This is more of a continuation of my earlier post on Applications of AI in different industries. I had not covered the retail industry as such, and there were a few enquiries from my friends on how the retail landscape is changing with AI. So, here’s an attempt at putting together some of the changes that we are getting to see around us.
But before delving into case studies and examples, let’s just remember that AI is no longer about smart robots walking around carrying things and doing odd jobs, just as retail is no longer about visiting the store and picking up products from shop shelves.
Both of these concepts have undergone major changes. And retail has integrated itself into our daily lives in a number of ways. Buying a product is no longer equated with making the Thanksgiving shopping trip or hitting the supermarket. Buying can happen from your social feed or when you are casually browsing the internet. This in turn has triggered off a series of solutions around visual commerce, recommendations, assistance in shopping, location hunting and others.
To start with, we have automation in analytics for segmentation, which is redefining and customizing the retail experience for customers.
Segmentation: If we consider the current state of retail to be Retail 4.0, then at the forefront of the change is automation for customer segmentation which is getting closer and narrower to the consumer with the technology of collaborative filtering. In layman’s terms collaborative filtering is a process of making automated predictions of the interest of a user by collecting preferences or taste information from many users. The underlying assumption is that if Person A has an opinion similar to Person B on one particular issue, then there are chances that Person A will have a similar opinion to Person B on other issues as well. These recommendations are targeted at a specific user with information gleaned from many users. The other differentiation that has been created through automation in analytics is creating a ‘segment of one’ i.e maintaining hyper granular behavioural profiles of shoppers at scale. For example, Amazon can track millions of shoppers and provide recommendations based on their shopping history thereby creating long term shoppers out of short term visitors.
Intelligent Assistants: The focus of AI has not only been hyper granular targeting but also addressing the ‘experience’ of commerce itself i.e redefining everything from the impulse to buy to the actual buying process. And Intelligent Assistants happen to be one of the solutions that retailers are experimenting with.
Macy’s for example is testing an AI based app called Macy’s On Call where the customers can ask ‘where is the shoe department’ and the app will point them out. The idea is not just to offer a helping hand at the store, but also to create an intelligent entity which will learn about the customer through multiple such interactions and offer him/her a truly enriching experience over a period of time.
Walmart is also partnering with Five Element Robotics for a new shopping cart called Dash which will help customers find the right items while relieving them of the strain of pushing a heavy cart. In fact, the payment can also be made through the cart.
The retail chain Lowe has also introduced its robotic assistant LoweBot in association with Fellow Robots. LoweBot is both a customer service as well as an inventory management assistant. he robot is capable of scanning inventory and saving relevant data. This data helps the retailer make quicker business decisions. The robot uses 3D scanning to detect the human frame and communicates with customers through speech recognition.
Retail Locations: Location matters in physical retail. Unless you are on the right street and with the right visibility, there’s a chance that people will miss your store. Hence we have solutions like SiteZeus. Pegged as a big data driven location intelligence platform, SiteZeus helps retailers identify the right location to set up their store.
Visual Search: There are tons of articles on how visual merchandising is shaping up with predictive analytics and how shoppers are being thronged with just the right t-shirt in just the right color. AI driven platforms like GoFind create the difference where you can click someone who’s wearing a tshirt you like and conduct a visual search of it on the platform. The AI searches through 1000+ online stores, across 100 million items, in seconds.
Dynamic Pricing: AI can be used for dynamic price-setting as well based on Neural Network Demand Models. For each customer, the optimum price can be calculated in real time, thereby significantly increasing the likelihood of a sale and this itself can change the entire dynamic of an industry that is quickly moving online. An example is the 5Analytics AI platform, where shops can integrate dynamic pricing into their existing systems via standard interfaces. Historical data for each customer is used to calculate his or her individual price range for specific product groups, and machine learning ensures this range is continually updated.
However, retail as an industry is not just dependent on pricing and shopping experiences. It is a large gamut with logistics, location, and a host of other aspects embedded in it. And the potential of AI across the various fields is immense. In some of the forthcoming articles, we will look at how AI is impacting areas like logistics as well.
