Putting AI back into Retail

The article explores the need to infuse Artificial Intelligence into the retail sector to optimize the marketing spend, improve the sourcing and fulfilment methods

We all have heard and read about the plethora of retail use cases using Artificial Intelligence (AI) in myriad ways - deep personalization for product recommendations, dynamic product pricing, using weather data to influence inventory, customer service improvements through chatbots and so on. While all these examples are legitimate and add value to retailers (and consumers), it is arguable whether these are truly AI applications.

I would argue that this first wave of so-called AI applications are better described as predictive analytics and optimization techniques that are purpose-built to address specific business challenges. There is nothing wrong with that and building these analytics into your business is certainly the right step forward to applying real AI in the future. Plus, given that most retailers still struggle with delivering a true omnichannel frictionless customer experience, thanks to all the disparate and often disconnected systems that exist today, these specific solutions start to bridge that gap and deliver benefit in the near-term. However, this then begs the question - what truly defines AI?

Artificial Intelligence can be defined as a science and a set of computational technologies that are inspired by the ways people use their nervous systems and bodies to sense, learn, reason, and take action (although in reality, they operate quite differently). In fact, an important distinction to know in the field of AI is between “narrow AI” and “general AI”. Narrow AI is defined as “a machine-based system designed to address a specific problem (such as playing chess)”. By contrast, general AI refers to machines with the ability to solve many different types of problems on their own, like humans can. Although general AI is currently a hot research topic, it is still likely decades away from true realization. And by that definition, to date, all applications of AI are examples of narrow AI.

Of course, there is a lot of promise even with the current advancements in ML, Deep Learning and other techniques. Facial recognition, natural language processing, self-driving vehicles and smart devices can all potentially work together to deliver a fantastic customer experience. Salesforce Einstein can pair that with automation and you have the insights and time to truly connect with your customers. Innovative retailers are already using such tools to optimize their marketing spend, improve their sourcing and fulfilment methods and infusing AI into everyday shopping experiences.

Having said that, AI is not just about developing smarter machines and solutions to make decisions faster and processes more efficient. The holy grail would be to use these advancements in a human environment - where men/women and machines are working side-by-side. Salesforce Einstein, for instance, learns from heaps of data collected,  embedded within the context of business, automatically discovering relevant insights, predicting future behaviour, proactively recommending best next actions and even automating tasks.

The notion of man and machine working together in ways that seem intelligent can be termed as Collective Intelligence. The prospect of Collective Intelligence in Retail, which is historically a people-heavy industry, is very exciting. Retailers continue to fail on delivering memorable customer experiences and deploying more technology or just replacing human labour is not the answer. Retailers need to figure out new ways to augment human empathy and creativity with machine intelligence to collectively deliver those memorable customer experiences.

While it’s easy to fall into the technology trap and mimic what others are doing, let’s face it - Amazon can do certain things because it is Amazon. They can operate an ultra-efficient robot-run warehouse supplying self-driving trucks which feed drones that deliver to your doorstep at the push of a button on your washing machine! Retailers will need to figure out what's good for them and more importantly, for their customers and apply AI techniques in a meaningful manner. In fact, one common misconception (thanks to mainstream media) is that AI will replace entire jobs or roles. On the contrary, the real value of AI is in replacing only specific task that are menial, routine, repetitive, needing analysis of massive amount of data and leaving the rest to the humans, who can bring the power of creativity, reasoning, storytelling, empathy and most importantly the subject matter expertise to serve customers better.

AI is a tool- a means to an end- increasingly embedded in everything for the benefit of the customer. If deployed correctly, it will change the face of Retail forever. And in the meantime, we have to be trusted stewards of data and data science and actively seek out opportunities delivering AI efficiently and invisibly. Retailers who recognize this early, make investments in retraining, reskilling and repositioning their people, design and develop experiences by effectively applying the insights from AI, will emerge winners.

Tags assigned to this article:
artificial intelligence retail ChatBot


Around The World