Clickstream and mobile app analytics, email campaigns, in-store purchases—all of these formerly disparate data sets have the potential to be combined into a single source of truth, a single view of your customers as they engage with you in increasingly complex ways. Modern retailers can leverage the power of big data and customer data management to create real-time, multifaceted data sets for email campaigns, online optimization, assortment planning, and more. A single view of the customer can be a very powerful tool to help distinguish a modern retailer, with huge benefits for both the business and the consumer.
Every customer engages with a retailer differently: viewing marketing emails over morning coffee, shopping for shoes online at lunchtime, or stopping in at a store on the weekend. Each of these touch points offers a retailer bits of information that can provide value if they’re harnessed and stored. And having a single view benefits your customers as well.
According to Eric Thorsen, vice president of industry solutions at Hortonworks, “The benefit to the consumer is what they’ve been asking for for decades. We can even go back a few hundred years, where you have the drugstore on Main Street, and the proprietor knows the population, he knows the community—someone comes in, he knows what they’ve purchased in the past, and he can recommend, ‘Hey, I just got a shipment of this. I think you might be interested.'” Knowing how your customers engage with your brand is pivotal to a retailer, and the solid foundation of this relationship is the single view of your customer.
Increasingly complex personalization strategies can help modern retailers offer contextual relevance to the shopping experience. With a single view of the customer, a retailer can know the brands a customer prefers, the styles they look at, and what other brands they aspire to interact with. What do you know about your customers? Is it cold where they live? Should you recommend purchasing a jacket or hiking boots? Answers to questions like these allow modern retailers to offer both the best products possible and the best shopping experience for their customers. According to Pitney Bowes, if a retailer fails to combine online and in-store behavior, it “has the potential to sully a customer’s perception of a brand because it illustrates a marketing disconnect not only between each of the retailer’s channels, but with how the consumer expects to be treated.”
Machine learning offers retailers next-generation capabilities. When asked how these work, Thorsen has the following to offer retailers: “[Algorithms] can start to sense and learn through pattern recognition. We can start to see zones and routes created through these stores, from a physical perspective—just like you can see somebody working through a website from a digital perspective. The concept of the path to purchase has been around a long time: What path does this consumer take as he or she makes a purchase? That’s a common phrase online—that’s also a common phrase in store. What’s brilliant is you can start to learn from both those environments.”
Machine learning algorithms offer retailers the ability to segment customers, predict next best offers and recommend items in increasingly sophisticated ways, but machine learning algorithms can fail to rise to the optimal level of intelligence if they’re not powered by single-view-of-the-customer data sets—it is big data that fuels these engines. Thorsen elaborated on the importance of personalization by writing, “There are two main places to upsell: during the path to purchase and at check-out. Both require personalized and relevant product placement. The consumer won’t buy knitting needles if they are looking for car parts. They won’t buy high-priced items if they are in a low-priced frame of mind.” In other words, you can’t just offer a good item for a customer to purchase: It must arrive at a time and with a context that is personally relevant—something that can only be achieved with a system for customer data management that provides a single, holistic view.
For developing a customer data management solution, there are battle-tested solutions and methodologies that have emerged as clear winners. Systems need to be able to tie previously siloed data sets into large, denormalized views of the customer. Data analysts can use these rich data sets to create increasingly complex reports and analyses of customer trends, and data scientists and machine learning researchers can generate mathematical models without having to be involved in data engineering tasks. Systems that are able to provide single views of the customer are rigid enough to power enterprise-level transformation and flexible enough to be tailored for individual use cases.
Does your system allow you to ingest weather data? How about economic forecasts? The key to success for these systems is the ability to evolve over time as needs change. Personalization and big data can help provide a single view of the customer—a strategy that is becoming critical for modern retailers to succeed in today’s omnichannel world.
To learn more about how retailers are moving beyond being reactive and becoming adaptive, check out this white paper.