It’s really no big secret. The key to a successful retail business, whether online or offline, hinges on only a few basic tenants: price, selection, in-stock, and of course personal and relevant customer engagement.
But anyone who has been in the retail industry for longer than about ten minutes knows that one can spend a lifetime trying to perfect this “retail holy grail,” only to eventually crash and burn.
That’s because doing all of these things well is almost always antithetical to having strong financial results. Each of the core retail financial metrics – revenue, margin and inventory turn – is negatively affected by having a large selection that is always in stock, offered at low prices. And the expense required to deliver personal and relevant interactions with customers can be devastating.
Delivering the perfect retail customer experience and making money at the same time simply cannot be done.
Well, one large online retailer has managed to come pretty close to the full trifecta. Even though they don’t have stores or salespeople, they carry more than 300 million SKUs, and can deliver any of them to their customers within 48 hours. They always respond to any customer interaction within seconds, and do whatever it takes to satisfy them. And don’t get me started on their prices.
How did these guys get to be so successful? One word: plastics. (No, actually the word is technology).
Here’s the irony: Our industry actually sells technology – flat-panel TVs, mobile phones, PCs and so on. And we can honestly go out on the limb and say that even though we are purveyors of high-tech products, we are way behind the eight ball when it comes to utilizing technology to run our retail point-of-sale (POS)) and enterprise resource planning systems (ERPs). We still manually forecast unit sales, randomly choose pricing, run shotgun sales with no customer segmentation, and so on.
As we like to say, when it comes to utilizing technology, the CE industry is the second to last, followed only by the music business.
So, let’s take a moment and define exactly what we mean by technology when it comes to retail and even wholesale systems. The modern-day translation is … artificial intelligence. That’s right: not flat screens or fast wireless networks, but instead software technology.
Online powerhouses have been deploying AI since 2007. And in the case of the “big dog,” their employees at this point do very little except pick stuff in the warehouse and tweak the algorithms to make them more exacting. It’s a fact that these guys spend more on technology than the next six-largest retailers combined, and arguably that alone is the secret to their success.
But now it’s your turn (which gets us to the whole point of this column): We want to show you how you can utilize the latest technologies and innovations to achieve more. We want to level the playing field so that all retailers – big and small, online and offline, A/V and IT – can all enjoy the same benefits of the AI revolution.
If you don’t know about the AI revolution, let us break it down for you: Artificial Intelligence is changing the way we conduct our business and live our lives faster than any other time in history.
What Is AI?
Yes it’s just like HAL in 2001. Robots will rule the Earth – well maybe sometime. But right now, let’s focus on a single aspect of AI called machine learning (ML).
Machine learning uses very complex computer programs called algorithms. These algorithms do what humans do – only much, much faster and way, way more accurately. And here’s the rub: they actually learn from their mistakes by training themselves, just like human beings. Some types of ML algorithms can recognize images, and some can recognize, speak or translate languages. Others can forecast customer demand, or even predict what customers will buy next.
Essentially, anything that a thinking human can do – reasoning, strategy, marketing, buying, hiring, training … anything at all – can be done faster and more accurately by an algorithm than by you.
Yes, dem’s fightin’ words to be sure. And you may not like it. But unfortunately, AI is here and is a key factor that separates the winners from the losers. And it is what has enabled successful technology companies including the large online guys to enjoy such all-encompassing success.
What it means for you is the ability to know what customers want before they tell you; the ability to create the assortment and inventory for your stores at a hyperlocal level to ensure that you have the perfect, relevant product mix for the customers in that market for that time of year. It also means that you are looking at significant performance gains because of reduced out-of-stocks and overstocks. Data and machine learning is enabling retailers to evolve and adapt to rapidly changing customer expectations. It is defining who will survive.
Beyond merchandising and inventory-control buying, machine learning can also be used across these other critical retail functions:
*online content/web design
*new store development/real estate
*new product development and private label
Future columns will eventually expand into wholesale supply chain and manufacturing. But honestly, retail has enough need for machine learning enhancements to keep us busy with this topic for a while.
ShiSh Shridhar has worked at Microsoft for more than 20 years, currently as retail industry lead for data and analytics. His partner Noah Herschman is a Microsoft retail industry senior architect with over 30 years’ experience in CE retail, including stints at Tweeter, Amazon, Staples China, eBay, DHgate and Groupon Goods Asia. Together they are creating technical solutions that are sophisticated in design and specifically targeted to improve businesses by engaging customers, empowering employees, optimizing operations and transforming products. You can reach Noah at firstname.lastname@example.org and ShiSh at Shish.Shridhar@microsoft.com.