Machine Learning And Artificial Intelligence: Creating A Retail Innovation Boom

0aGary Hawkins CARTArtificial intelligence (AI), boosted by machine learning, is feeding off the massive data available across the retail industry to transform everything from manufacturing to marketing. Amazon uses AI to make product recommendations, Netflix’s AI knows what you’ll want to binge and The North Face uses IBM’s Watson AI system to power natural language recommendations.

According to recent report from research firm Markets and Markets, the artificial intelligence market is expected to grow to $5.05 Billion by 2020, as a result of the adoption of AI in the media and advertising, retail, finance and health care sectors. And while AI is not new, these gains are being powered by the exponential growth of low cost computer processing power and new approaches to ‘teaching’ computers how to make decisions. AI, and associated machine learning, is feeding off big data to deliver cloud-based capabilities that are transforming the retail industry.

One such area that is being heavily impacted by AI in retail is the supply chain. Cutting edge solutions are revolutionizing the retail supply chain from procurement and production through customer delivery. Just a few examples of these technologies include:


Robotics: Robotics is transforming distribution center operations (DC), improving order picking accuracy and speed. According in Gartner, 45% of the fastest-growing companies will have fewer employees than robots by 2018 and Amazon is the best example of this. The company has publicly stated that the use of robotics in a growing number of its DCs is helping the company lower operating costs by 20%. C&S Wholesale Grocery is similarly employing robots in its warehouses to pick orders. Walmart is using drones within its massive warehouses to monitor out-of-stocks and slotting accuracy. And one day, self-driving cars may be employed to deliver goods to the customer’s doorstep.

Optimization: AI and machine learning are used to optimize distribution center layout, along with picking and delivery schedules. Sophisticated capabilities optimize efficiency throughout the entire warehouse-to-store process, picking orders in the warehouse and building pallets for each aisle of the store.

Replenishment: Much has been written about Amazon’s Dash Replenishment Service being integrated into a growing number of home appliances, bestowing upon the dishwasher (for example) the ability to know when to automatically reorder detergent. AI services such as these are already seeing massive market adoption, with Gartner expecting to see 6.4 billion connected products in use worldwide and $235 billion in connected services in 2016 alone.

Segment-Of-One Marketing

While AI is transforming supply chain activity, AI is also poised to cause major disruption in marketing and merchandising.

Many of the largest retailers leverage customer intelligence — customer identified purchase data, demographic information and other attributes — to power customer-centric initiatives. There is no question about the benefit of these programs; one has to only look at Kroger’s 50 consecutive quarters of same store sales growth to see the impact of its Customer First strategy.

Behind these efforts are expensive analytic consulting firms employing armies of data analysts to crunch through massive amounts of data to create a growing number of shopper segments, analytics and insights that in turn feed marketing personalization, product assortment and promotion planning for their retail customers. While powerful, few companies can afford the massive outlays to begin the cycle and fewer still have the operational discipline to succeed with such a strategy.

But AI is changing this, making sophisticated marketing personalization available to even the smallest retailers. New entrants are coming into the market using AI and machine learning to power hyper-personalization and delivering it as a cloud-based solution to retailers of any size.

Further, targeting is done at a true segment-of-one level — the individual customer — leveraging massive amounts of purchase data, calculated brand loyalty, discount propensity, purchase frequency and other categorical trends to suggest the right promotion to the right customer. The platform makes use of machine learning to automatically ‘learn’ from each promotion sent to a customer and uses that learning in the next iteration.

AI-powered marketing is poised to disrupt decades of mass trade promotion and shopper marketing initiatives as systems gain innate intelligence to power strategic personalization and personalized pricing.

Merchandising With AI

On the merchandising front, product assortment optimization represents a massive opportunity for retailers. The power of having the right products on the shelf in each store based on the customers of that store reduces out-of-stocks, increases sales and provides an improved quality of shopper visit. Again, the only path to this was an expensive analytics approach such as that used by Kroger and a small number of other big retailers…until now.

One such solution employs the latest AI and machine learning technologies to the merchandising challenge. The solution not only ingests any customer intelligence (segments, etc.) that the retailer may possess, but also factors in extensive knowledge of in-store and out-of-store promotional activity to optimize product assortment within each store. These sophisticated capabilities are available to even smaller retailers as a cloud-based solution.

As next-generation AI-powered solutions across retail, they will connect with other solutions to turbocharge the impact. Many retailers today have digital cameras in place to count customers entering and exiting the store. A growing number of retailers have either digital cameras that expand this customer counting to the departments and aisles or use WiFi or beacons to provide similar data. Some solutions even use anonymous facial recognition to report the mood of shoppers; are they angry, happy, frustrated and so on.

Now connect these different capabilities in real time. Once the retailer knows customer traffic throughout the store every hour of every day, they can take action on that knowledge. AI and machine learning is capable of distributing specific messaging and offers to each individual customer entering the store to ensure maximum aisle conversion, or, simply put, what percentage of total shoppers you can entice to go down a specific aisle. This is not science fiction; this capability creates a new front in retail competition and it is here today.

Artificial intelligence and machine learning are rapidly becoming embedded in services and solutions across retail and are affecting cost efficiencies, improved forecasting and marketing personalization. Technology-fueled innovation is growing at an ever-increasing pace and artificial intelligence will increasingly transform and disrupt the retail industry.


As Founder and CEO of CART, Gary Hawkins has deep insight into current and future innovation in fast moving consumer goods retail. Hawkins reviews thousands of new solutions each year and has over 30 years of industry experience leading shopper-focused innovation across the supply chain. Prior to founding CART, Hawkins played an active role working with M&M Meat Shops (Canada), Mitsubishi (Japan), ASDA (UK), Procter & Gamble, Unilever and other prominent companies in markets around the world. He has founded or been strategically involved with numerous successful startups including Convena, RetailNext, and Retalix (acquired by NCR). Earlier in his career, Hawkins was CEO of Green Hills, called “The Best Little Grocery Store in America” by Inc. magazine.


Feature Your Byline

Submit an Executive ViewPoints.

Featured Event

Join the retail community as we come together for three days of strategic sessions, meaningful off-site networking events and interactive learning experiences.


Access The Media Kit


Access Our Editorial Calendar

If you are downloading this on behalf of a client, please provide the company name and website information below: