Imagine you’re about to leave the house to pick up your kids. As you grab your keys, you hear a voice from the device on your coffee table: “It looks like you’ll use the last of your milk tomorrow, and yogurt is on sale for $1.19. Would you like to pick up an order from Trader Joe’s, for a total of $5.35?” You say yes, and Alexa confirms. The order will be ready for curbside pickup, on the way home from your kids’ school, in 15 minutes.
This future scenario isn’t so far off. Amazon, Facebook, Google, and Apple are accelerating consumer expectations and what’s technologically possible, from same-day delivery to machine-powered image recognition. You can call an Uber with Siri and book a flight entirely through a Facebook Messenger bot.
Responsive retail has peaked, and we’re about to enter the era of predictive commerce. It’s time for retailers to help people find products in their precise moment of need — and perhaps before they even perceive that need — whether or not they’re logged in or ready to click a “buy” button on a screen. This shift will require designing experiences that merge an understanding of human behavior with large-scale automation and data integration.
Machine Learning Beyond Forecasting
Retail giants have been using machine-learning algorithms to forecast demand and set prices for years. Amazon patented predictive stocking in 2014, and saying that AI, machine learning, and personalization technologies have improved since then is an understatement. Retailers need to think more like tech companies, using AI and machine learning not just to predict how to stock stores and staff shifts but also to dynamically recommend products and set prices that appeal to individual consumers.
Say you’re on a business trip and realize you forgot your phone charger. You’ll pay a premium for a new one delivered to your hotel room before an all-day meeting. An electronics retailer might also predict that you want new headphones. It can offer you a deal on a noise-canceling pair at a price that accounts for current pricing on Amazon, in-store inventory at Best Buy, the current rates for on-demand couriers, and the fact that you’re taking a red-eye flight home tomorrow.
This level of prediction requires detecting subtle patterns from massive data sets that are constantly in flux: consumers’ purchase histories, product preferences, and schedules; competitors’ pricing and inventory; and current and forecasted product demand. This is where AI and machine learning comes in and where companies are investing. Last year, Etsy acquired a company that specializes in machine learning to make its searches more predictive by surfacing nuanced product recommendations that go beyond simple purchase histories or preferences. This is the natural evolution of product recommendations, one that will be the standard for years to come.
Retailers need to think more like tech companies, using AI and machine learning not just to predict how to stock stores and staff shifts but also to dynamically recommend products and set prices that appeal to individual consumers.
Realizing the Potential of Connected Devices and Data
Predictive retail involves inspiring consumers in different contexts — before, during, and after a purchase. Commerce is already becoming less of a deliberate activity than an organic part of how we experience daily life. It’s not just smartphones that make browsing and buying spontaneous; Amazon’s Alexa-powered Echo device and Dash buttons are enabling purchases in the home. You can hit the Tide Dash button in your laundry room when you see that you’re running low on detergent, or ask Alexa to order your mom a bouquet of flowers when you remember that her birthday is next week. This is just the beginning.
The next generation of smart assistants and connected devices will learn from user habits and pick up on behavioral and environmental patterns in order to make these experiences more predictive. Devices like the Echo will access data from everyday interactions to predict specific opportunities for a transaction.
There’s also huge potential for connected devices in retail stores to predict consumer behavior and respond to individual needs. Many stores are already using smartphones to follow customers’ activity and deliver context-specific offers. It’s not a stretch to imagine that the evolution of biometrics, identity technologies, and location sensors will allow retailers to personalize content based on factors such as how you’re feeling, how much time you have to browse, and whether you’re coming from the office or you’ve just finished working out.
Retailers will need to program brick-and-mortar experiences with the same targeting and personalization they offer online. Think about walking past Nordstrom and receiving a notification for an offer on a new pair of sneakers. Your current pair is worn down from running almost 500 miles — all logged by a chip in the sole that sends data to your fitness app. You swipe the notification to select the styles you want to try on, and an in-store map guides you to an associate waiting with your shoes.
Embracing Human-Centered Design
The future of predictive retail requires designing new ecosystems for commerce. These systems will be built around the human, rather than around a particular device or around online or offline experience. These systems will need to incorporate human connection and storytelling, spatial design and context, and a lot of data.
Many retailers are getting ahead of this shift by creating innovation labs — teams and spaces dedicated to incubating new ideas and testing digital experiences that connect the online and in-store worlds. Sephora’s Innovation Lab is a great example. The brand introduced a “store mode” for its mobile app, which integrates a user’s online shopping cart and Beauty Insider loyalty card to remind them of the products they’ve saved, the points they’ve earned, and the benefits available to them, such as a free makeover.
Retail chains, brands, and e-commerce companies are also collaborating to bring new ideas to life. Several years ago, Westfield Malls’ lab worked with eBay to build 10-foot-tall interactive screens in its San Francisco shopping center. Shoppers swiped these screens to browse products from brands like Rebecca Minkoff and Sony, which they could purchase directly on mobile.
There’s huge potential to layer predictive capabilities on top of this AI-driven infrastructure. Imagine a store window that connects with your phone to display personalized content. For instance, you might see gifts for your partner’s birthday or swimsuits for your next vacation, customized based on the boards you follow on Pinterest and the brands you follow on Instagram. By connecting data from multiple sources and designing for the user, retailers can create more-relevant experiences that pull you into a store, website, or app. Even more powerful, they can predict what you want before you do.
These systems will need to incorporate human connection and storytelling, spatial design and context, and a lot of data.
Considering Privacy, Building Trust
There’s almost always some trade-off between privacy and personalization; this has been true for every generation of technology. Retailers need to move forward with transparency, respect, and security as their priorities. They also need to show value. Google has done this well, not just with personalized search results but also with services such as Google Now, which integrates with your calendar, and Google Maps to alert you that traffic to your meeting is heavier than usual and tells you when you should leave the office to arrive on time.
Many of us are inclined to share personal information for experiences that are magical and valuable — and that we can’t get elsewhere. Retailers will need to create experiences that make this magic and value apparent. The revolution is already under way. Tomorrow, people will expect even faster and more-intelligent service than they do today. At a point in the very near future, the expectation will shift from on-demand to predictive commerce. It’s time for retailers to get ahead of that change.
This article originally appeared on HBR.org.