Johannesburg, 07 Aug 2017
The impact of emerging technologies such as artificial intelligence (AI), machine learning and cognitive computing - the latter underpinned by big data and advanced data analytics - is beginning to be felt. Although still in its infancy in South Africa, the manner in which these technologies are being adopted in the developed world suggests that it won't be too long before the retail experience of the future is upon us.
At present, the major challenges facing local retailers, in terms of implementing these technologies, are skills shortages, poor data quality, challenges with analytical modelling and the fact that the analytical value chain has not yet been fully automated and remains manual in a lot of respects.
According to David Cosgrave, Customer Intelligence Lead at SAS, there also seems to be a lack of commitment at board level, which means data scientists are seldom used properly. Add to this the fact that there is an element of risk taking and the potential for failure and it is no surprise why SA still lags behind some of the more developed nations in respect of the adoption of these technologies.
"At the same time, the benefits of being a successful early adopter are huge, as these technologies will enable retailers to gain a substantial competitive advantage, massively increase efficiencies and earn significant cost savings - which can then be passed on to the customers," he says.
"The gains provided by technologies like advanced analytics and AI are such that it will enable large retailers to truly optimise both their stores and their supply chain, allowing them to have the right mix of products, for the right customers, at the right store, at the right price, at the right time."
In terms of supply chain management, think about how simple this technology could make it to, for example, plot a delivery route that only uses left turns. In this way, he continues, the delivery vehicle does not have to turn across traffic, which leads to improved fuel costs and time efficiencies. Reducing supply chain and delivery costs means the retailer can reduce prices, which will naturally attract more customers.
"Moreover, this is just the start. As AI becomes more pervasive and is used in more areas, so the ecosystem becomes bigger and the benefits it offers businesses become greater. We are not that far from having self-driving vehicles, thanks to AI. Now think of the route example outlined above and consider that the delivery truck is driven by an AI - it will mean the vehicle itself is driven at maximum efficiency, reducing fuel consumption and wear and tear, reducing the retailer's costs further."
"In-store it will be the same, with AI and analytics working together to reduce the pain of shopping, by eliminating queues, enabling the shopper to find what they want quickly and ensuring that stock is optimised so the store has what the client wants."
Cosgrave explains that an example of how a future, optimised store would work is that facial recognition technology detects when the consumer walks through the doors. The retailer already has an idea of the type of clothing they like based on past purchases and the parameters the customer has provided them with, such as height, weight, etc. The client stands in front of a mirror and AI technology immediately 'dresses' them in an outfit based on their preferences and unique style. The customer can easily flip through colours and styles without once having to visit the dressing room, or physically having to change clothes.
"Once the purchase decision is made, a robot brings them their purchase and they walk out the store without having to queue to pay, as the retailer already has their card details and the account is automatically debited for the purchase. This is the real future of retail: instant gratification on a whole new level."
The next evolutionary step from here, he adds, will be for the machines themselves to begin asking the important business questions.
"Perhaps the ultimate destination for AI in retail is for the machine to begin asking critical questions like 'how can the business improve its profitability?' In other words, instead of an analyst testing various hypotheses to determine optimal outcome, the machine does it automatically. Obviously, it does it better because it can rapidly process thousands of possibilities. It then identifies the problem, analyses the reasons why this is occurring and potentially even implements measures to rectify this - all without a human being involved at all."
"While the above possibilities remain some way off, what is important for businesses to understand is that, as with other now-ubiquitous technologies, AI will eventually be embraced by everyone. This includes your competitors, so those who fail to embrace it early enough will find themselves at a distinct disadvantage in the technological retail future that is just around the corner," he concludes.
Share