5 Ways Retailers Can Evolve Beyond Traditional Segmentation Methods

Today, consumers are interacting and engaging with brands in many different ways, especially as the dynamics of ecommerce have shifted over the past two years. For many retailers that are managing disruption in the current environment, every day is like Black Friday, with new customers they know nothing about and current customers who are shopping differently across and within categories. Retailers need to spend wisely for new customer acquisition and, since acquisition is often more costly than retention, they need to incorporate intelligent methods to build lifetime loyalty. Retailers should look ahead to the next generation of the digital experience, try to understand how the customer journey has changed, and move past mere personalization to build more empathy into the process.

That was the focus of a recent eMarketer Tech-Talk featuring the visionary founders of three technology companies, all of whom are Google Cloud partners that are focused on helping retailers activate and harness their data better: Fayez Mohamood, co-founder and CEO of Bluecore; James McDermott, co-founder and CEO of Lytics; Mario Ciabarra, founder and CEO of Quantum Metric. Our colleague Carrie Tharp, VP of Retail and Consumer at Google Cloud, moderated the conversation. Here are five key takeaways.

1. Think beyond customer segmentation and demographics. Topping the list is the drive toward using data and technology to change the customer experience as it’s happening. “Customers don’t go on linear journeys; they live in micro moments,” Mohamood from Bluecore commented. They interact with the brand across multiple touch points. Retailers need to understand that and engage customers in the moment through personalization and recommendations. That is a key area that Bluecore, an AI-driven retail marketing platform, is helping retailers focus on: activating product catalog and customer data across digital channels to drive repeat purchases and grow revenue. “It’s time to think past the classic focus on customer segmentation and demographics,” Tharp added. The trick is to understand people as individuals whose purpose in their different interactions with a brand varies wildly.

2. Democratize data so that those who are designing the customer experiences can use it in real time. With the advent of the cloud-based customer data platform (CDP), retailers can more quickly build and act on customer profiles using different types of intelligence relative to the customer’s intent and interests. “That’s the starting point that enables us to listen and understand their real-time behavior,” McDermott from Lytics advised. “But building profiles is not the end goal; how will you act on that data? How are you making the experience better for your customers?” McDermott co-founded Lytics, which focuses on behavioral and intent-based analytics, after discovering a need in the market to bridge the gap between data and action. He suggested beginning with the end in mind, with a use case to illustrate.

Doing that isn’t simple. Deciding which data has value and how to use it is the crux of the challenge. The good news is that the CDP has made more data accessible to data scientists, who are using it to build models that can be turned into systems of engagement. Internal teams are experimenting and testing continuously, with a rich set of hypotheses to figure out how to prioritize the signals and determine which experiences will actually drive better interactions. However, as Mohamood observed, “That’s only one piece of the puzzle, because data scientists are typically not actually creating the experiences that customers have on channels – whether it’s websites, email, SMS, and so forth. We need to make data democratic to marketers and business users who need it in real time to test and learn.”

3. Leverage artificial intelligence (AI) to anticipate behavioral and inventory changes. Rich product data and attribution are as important as customer data, Mohamood remarked. “Shoppers change, and then the data changes, and then inventory changes, and the data changes yet again. Your team has to be nimble to react to those signals.” That’s where AI comes in – augmented with human insight and intervention.

Some retailers have a constantly changing inventory mix, and can match what’s in stock to the customer intent. They adjust their messages to the customer accordingly, without changing the frequency of communications, thereby improving repeat purchase rates. Ciabarra from Quantum Metric also talked about the importance of AI to react to changes faster. “I know it sounds crazy, but real time really means real time. To allow me to go and ask questions of the data that are really complex and be able to come out with answers and build that analysis into the product, I think we’re going to see that unlock of AI and ML.” Quantum Metric’s focus with continuous product design is all about building analysis into and across the digital product design process so that teams can iterate faster and deliver the products that customers truly want.

In addition, Tharp pointed out that many retailers are adding marketplace capabilities to their sites, evolving from static inventory levels to “an expansive, endless aisle.” However, they are struggling to get the right operating model in place across marketing, analytic, and digital teams. How can retailers think about their operating model related to personalization and data?

4. Build the link between customer data and empathy. In short, the operating model has to support the full connection of all relevant interactions – for example, combining past purchase data with intent data, browsing data, and mobile app data. That requires the ability to integrate multiple data sets and operationalize higher-performing personalization. The technology stack is being adapted to be oriented around the customer, McDermott said, with the architecture becoming simpler and more centralized in the cloud data warehouse. One clear trend is that the CDP has become the central repository for customer data, with more intelligence built in using AI and machine learning.

A key goal is to create a more “one-to-one” connection with customers by enabling retailers to gain real-time visibility into the customer’s selection process and how they are doing in the checkout. “We have to have that level of empathy. I like to use the words ‘quantified empathy’,” Ciabarra said. “In a store we had that one-on-one connection, but online, it could be one marketer to a million customers. How do you get real-time visibility across the organization so they can consume and act on it? Everyone has to have a shared view.”

5. Transform the organization and break down silos. Just about every company today is focused on becoming more customer-centric. But the great majority are structured around specialized teams that function in silos with their own perspectives and maybe even different objectives. “Data tends to be held hostage by a few folks in the organization,” Mohamood noted. Enabling a shared view across the organization among marketers, merchandising, and digital teams requires a gradual cultural shift – in the best case, with executive buy-in.

Some companies promote collaboration by assembling pods that include marketers, data scientists, analysts, and creative people testing and learning together. Another promising development: marketers and data scientists are proactively partnering with finance and operations teams, especially around KPI design. They are working together to design metrics that are customer focused, rather than channel-focused, and reexamining the technology stack from that point of view. The mindset is evolving toward looking at data not from the perspective of marketing or product or UX and design, but through the customer lens, pulling the data together in a way that’s more usable.

We invite you to watch the full webinar to derive even more insights from this thoughtful discussion. In addition, learn more about Bluecore, Lytics, and Quantum Metric solutions all available on the Google Cloud Marketplace.