As retail evolves, 5G and edge computing keep you in the express line

As retail evolves, 5G and edge computing keep you in the express line

Originally published by CHRIS ANGELINI

This article is part of the Technology Insight series, made possible with funding from Intel.

Today’s customer doesn’t want to walk into a store and be approached by a clueless salesperson. They want efficiency. Their expectations are higher than ever. Plus, patience is at an all-time low. Those are daunting challenges for any retailer. Fortunately, troves of data promise to bridge the worlds of online and in-person shopping, creating engaging experiences.

As 5G, edge computing, and AI proliferate, we’re going to start seeing innovation that makes retail even more exciting. Real-time analytics, high-speed connectivity, and low latency come together to breathe life into personalized recommendations, augmented reality, video chat with subject matter experts, and more.

Key points

  • Today’s retail experience is efficient but impersonal
  • Data makes it possible to know what customers want before they step foot in a store
  • Retailers must master omnichannel — integrating mobile apps, social media, in-store shopping, and more — to create the seamless experiences
  • 5G, edge computing, AI, and IoT all have played key roles in the data-driven evolution.

Why do we need a better shopping experience?

According to research published by McKinsey ahead of the 2019 holiday season, 62% of shoppers planned to make their purchases online and in-store. Only 12% admitted to buying gifts spontaneously — in-depth research was far more prevalent. And the top motivator for participating in a shopping event was attractive offers. The winners, McKinsey concluded, would be retailers able to target their marketing to make it relevant, master the omnichannel experience, and win consideration with perfectly timed campaigns.

Adding computer vision to self-service kiosks makes them much more capable, enabling improved loss prevention, gesture recognition, and personalized offers.

Adding computer vision to self-service kiosks makes them much more capable, enabling improved loss prevention, gesture recognition, and personalized offers. Easier said than done, right? Data is what makes each of those goals attainable, determining who should see what, and when. Too often, though, technology is used for the sole purpose of improving efficiency, sacrificing intimacy in the process. Retail is bigger and faster as a result. But we’ve lost the personal touch of a real human who knows your face. While most shoppers believe that today’s self-service technologies improve the retail experience, those same tools aren’t very good at helping customers find products or make suggestions.

The future of retail, then, combines efficiency and personalization. It leverages data from sensors and analytics performed at the edge to create a more engaging shopping experience.

How do 5G and edge computing enable next-gen retail?

Let’s make this a little more real. You’re in the market for a new laptop. You read the reviews and have a couple of models in mind, so you throw them into a cart on your favorite electronics store’s website. But before you pull the trigger on one of them, you want to go hands-on. As you walk up to the door, facial recognition software that you agreed to use (trust is going to be a major topic of conversation here) identifies your face. It sends information to someone inside who’s ready to help find those machines. You make your choice and start heading for the door. But before you get there, a notification alerts you to a sale on wireless gaming mice and headsets. Nice save. With all three items in your arms, you walk back to the car. Your credit card is on-file, and sensors already scanned your purchases on the way out.

Before, during, and after your transaction, data is gathered, stored, and analyzed to create a seamless experience. Scaled out to hundreds or thousands of customers interacting with an even greater number of IoT sensors, that’s a potential mess without the right technologies in place. But 5G and edge computing come together and alleviate the bottlenecks imposed by previous-gen standards, widening the pipes in dense environments for information to flow in real-time.

Vispera ShelfSight employs IoT cameras, edge computing, and AI to monitor and manage shelf space in real-time.

Above: Vispera ShelfSight employs IoT cameras, edge computing, and AI to monitor and manage shelf space in real-time. Image Credit: Intel

Elements of this scenario are already on display. For instance, Intel hosted UST Global and Cloudpick in its booth at NFR 2020. Their Frictionless Checkout Store solution employs AI technologies based on OpenVINO, IoT sensors, and edge computing to identify products and shopping behaviors. With a retailer’s app running on your mobile device, you can walk into a frictionless store, walk out with a shopping cart full of goods, and automatically pay as you exit. Logjams at the checkout counter become a thing of the past, and associates are freed up to help with customer service.

Even self-service kiosks are learning new tricks with the help of AI and edge computing. Infusing self-checkout systems with computer vision, for instance, gives them the ability to confirm that the item you scan matches what’s in your bag. Vision algorithms have other applications, too. Identifying faces (to authenticate payment), recognizing gestures (for touchless commands), and facilitating personalized offers are all potential additions to the machines we use today.

As 5G proliferates, so will the opportunities to make retail sing. Virtual fitting rooms will leverage the high bandwidth and low latency of 5G networks to render customers clothed in the latest fashions, while compute power on the edge recommends complementary accessories. Smart shelves will help manage inventory. And data-hungry robots will help customers find what they’re looking for. All of those technologies will lead to better shopping experiences. “A positive experience can turn one-time guests into loyal guests, but it often requires a network with high reliability and low latency,” said Phillip Hartfield, GM of AT&T business solutions for retail.

The whole is bigger than the sum of its parts

In a recent blog post, Joe Jensen, general manager of Intel’s retail solutions division, answered the question: How will using data at the edge alter the retail industry? “It will enable retailers to take advantage of advancements in AI, computer vision, machine learning, augmented reality, IoT, and robotics. The benefits include becoming more responsive, agile, and customer-focused.”

Computer vision, IoT sensors, and deep learning come together in Amazon’s checkout-free Go stores.

Above: Computer vision, IoT sensors, and deep learning come together in Amazon’s checkout-free Go stores. Image Credit: Jordan Stead / Amazon

Surprisingly, or perhaps not, Amazon is stepping up to show more traditional retailers how it’s done. The company’s stores in Chicago, New York, San Francisco, and Seattle serve as the most recent examples of retail done right. Amazon is using computer vision, an amalgamation of sensors, and deep learning to enable checkout-free shopping. It’s also demonstrating a mastery of omnichannel by populating its Go app with inventory information from each location, allowing you to browse what’s available before visiting. Because the experience is so streamlined, the staff you find working can dedicate themselves to preparing food, stocking shelves, and answering questions.

Flashy new online and AR/VR shopping schemes may grab headlines. But by 2023, eCommerce is expected to account for just 21% of total retail sales, and a mere 5% of grocery sales. Although we’re all eagerly anticipating same-day drone-based delivery services, it’s clear that the experience of purchasing our favorite goods in-person isn’t going away. But shopping will definitely change thanks to technologies designed to ingest data quickly, process it at the edge, cough up analytics in real-time, and provide personalization like we’ve never seen.