How Machine Learning and Predictive Analytics Drive Today’s Retail Personalization

When something changes in the customer landscape, Walmart knows. And they know just how to react.

“Walmart has a massive inventory with millions of products,” says John Bates, senior product manager for data science and predictive marketing solutions at Adobe. “And they adjust that inventory to better align with certain types of products, depending on what’s happening in real time.”

For example, if a hurricane is in the weather forecast, Walmart will shift its inventory to have the things they know from past experience their customers will want to buy — extra grocery staples, bottled water, sandbags, wet/dry vacuums, chainsaws, and generators. Simultaneously, merchandise that is less likely to sell in this weather — again, according to Walmart’s data — is taken off the shelves.

“This strategy provides sufficient inventory for the most-needed items on any given day and minimizes the shelf time of all products — satisfying both customer and retailer needs,” says John.

Ensuring a relevant experience for customers, whether they’re heading to a store or shopping online, is achieved by leveraging the power of artificial intelligence (AI), including machine learning and predictive analytics, to deliver personalized experiences at scale.

AI Helps Deliver What Customers Want
Not surprisingly, retail and ecommerce have always been central to the personalization and optimization conversation. From Amazon’s recommendations — which drive 30 percent of its revenue — to targeted email outreach and push alerts promoting complementary products, the most optimization-focused retailers have always pushed the experience envelope, fueling people’s desire for more relevance at all touchpoints.

Delivering relevance on those touchpoints, though, is where some retailers start to lose their footing. “Taking that next step is a big leap,” says Kevin Lindsay, director of product marketing for Adobe Target. “It’s a leap of faith in terms of how much you can bite off. How much is actually doable today and what benefits can you get from incorporating AI into developing these tactics today?”

But delivering personalized experiences at every touchpoint isn’t something customers just want, it’s what they expect. More than half of consumers want a “totally personalized experience,” and three in five are happy to have interests and behaviors shared if it means a more personalized journey with a retailer. However, 42 percent of retailers say they know too little to effectively engage key segments.

Even a Little AI Can Help Deliver the Right Experience
Working with AI, predictive analytics, and machine learning perhaps seems out of reach for many retailers, however, as Kevin mentioned, it’s not an all-or-none proposition. Retailers that take a phased approach to implementing and applying the insights they gain from AI are the ones that are already benefiting. Think about how you can start applying AI to help you in each of these areas.

Invest in the right technology stack. Because many retailers haven’t made the leap of faith to invest in the right technology stack that delivers relevance at scale, the experiences they deliver are more likely to miss the customer experience mark. From ecommerce experiences to connected store associates to post-sales communications, without the machine anticipating next steps by acting on predictive analytics, retailers can’t effectively and efficiently map out the customer journey — and, naturally, can’t act on those critical cues and moments in time. Start by taking inventory of the data your organization has access to and how it is integrated for a complete view of your customers.

Surface customer needs. Retailers also aren’t able to leverage key data points and real-time actions to deliver relevance beyond what’s right in front of them. “There are plenty of other applications that come along with machine learning,” John adds. “ Discoverability of content in search is a good example. By leveraging machine learning and predictive analytics, brands can look beyond what customers are searching for and start connecting the dots on what they likely want — it’s cross-selling at scale, matching customers to specific products or content that will nudge them towards more conversions and greater lifetime values.”

ASOS.com, a British online fashion and beauty store, uses AI to uncover and solve issues specific to online retailers — helping customers find the right size and minimizing returns. By analyzing which items customers keep, in which sizes, versus the items and sizes that get returned most often, ASOS is able to use machine learning to recommend appropriate sizes for individual customers regardless of the brand or fit of specific items of clothing. As a result, returns of ill-fitting clothing are minimized, the customer experience is improved, and ASOS reduces its costs.

Produce relevant cross-channel interactions. When machine learning and predictive analysis do take the wheel, cross-channel customer interactions become increasingly relevant to customers on an individual level. And that surprises and delights those consumers at every turn, and all but ensures they keep coming back for more. Says John, “The impact is very straightforward. Machine learning and predictive analytics increase the likelihood a customer will convert — or, even decreases the likelihood an undesirable outcome will occur. That could be something like low retention for a subscription service.”

Gather more data. Retailers should act on every opportunity to gather data. “Every single point of interaction that a consumer has with a retailer is another dot. It is another piece of data that helps to make up the picture,” explains Kevin. The picture you create with data ultimately will feed machine learning and predictive analysis for retailers. Brands like The Home Depot and Ikea are good examples of companies moving on this data, as they’re using beacon technology to understand the physical journeys and pathways that people take within a large mass merchant store. And the data that emerges provides an interesting insight into how they should be merchandising their products.

Incorporating AI is a shift that’s happening daily but, for most retailers, isn’t quite there — yet. “The ability to say, ‘OK, here is everything we’re learning,’ and then ask how we can act upon it right now to provide a customer with a much more relevant experience — I would say that is the piece that is not very mature yet even among bigger retailers,” says Kevin. “You can probably count on two hands the number of big retail companies out there that have the data, resources, and ability to build machine learning systems for the benefit of personalization.” Start small, but start, and you’ll be at the top of the pack when it comes to delivering personal and relevant experiences across your customer base.

The Future of AI in Retail Experiences
The technology powering artificial intelligence is quickly growing and evolving. “There’s a lot more we’ll see,” John says. “More intelligent systems with cognitive analytics — systems that go beyond serving up insights to actually make recommendations and decisions based on those insights, and then constantly learn to make better decisions.”

Investments in AI at Adobe are consolidated under a single framework with Adobe Sensei. Sensei will unify AI components along with trillions of data and content points to create unparalleled experiences. In Adobe Target, a new experience decision engine dubbed One-Click Personalization is now in beta and enables marketers to test different web page layouts and activate the process with a single click. After that, the machine takes over, working through several hundreds of thousands of visits and interactions with the website to determine the ideal layout — the one that drives the most conversions.”

And that’s just the beginning. Take steps now to incorporate the power of AI in your efforts to drive personalized and relevant experiences to each of your customers.

For more insights on how retailers are adopting new technologies for more personal customer experiences, read more from our digital marketing retail series.

And, download our white paper to learn why retailers that use experiences stand out.

The post How Machine Learning and Predictive Analytics Drive Today’s Retail Personalization appeared first on Digital Marketing Blog by Adobe.

from Digital Marketing Blog by Adobe https://blogs.adobe.com/digitalmarketing/analytics/machine-learning-predictive-analytics-drive-todays-retail-personalization/

via Tumblr http://euro3plast-fr.tumblr.com/post/162117632434