Long before digital marketing, marketers focused on the question of “who” someone was to a brand. The primary way to define “who” was through recency, frequency and monetary (RFM) analysis. Ironically, as marketers moved into the digital space, the basic concept of “who” was largely discarded, replaced by “what” (i.e. “What items did a shopper look at?”; “What did they add to their cart”). This, in turn, led to a focus on open rates, click-throughs and conversions.
Today, thanks to the democratization of big data, there is renewed focus on “who.” Retailers of any size are able to gather more types and a greater volume of data than ever before. But merely collecting data doesn’t solve the problem. Retailers large and small continue to struggle to take data from intake to insight. In fact, Accenture reports that 58 percent of retailers surveyed admit that harvesting customer insight from collected data remains a key challenge. That’s where customer intelligence comes in.
In the simplest terms, customer intelligence is the practice of identifying and differentiating customers into segments using data-driven methods. In order for retailers to put successful customer intelligence into practice, they must first have the right team in place. We’ve determined it takes the right alignment of three key elements: 1) retail experts, 2) data scientists, and 3) predictive modeling experts.
Retail experts are needed to frame a problem properly in a way that will provide value to the brand. Without the retail expert’s industry experience, the data experts and predictive analytics gurus may be able to build excellent models that produce stunning accuracy, but will also answer the wrong questions.
Data scientists are needed to identify data available for predictive modeling and decipher how that data can be accessed and normalized. These vital team members are chiefly responsible for identifying rich data sources, connecting them to online and offline data sources and drawing insight from the resulting data set. Without data scientists, retailers would fall into the trap of building one-off systems that silo data so it cannot be shared, retrieved or made useful later.
Predictive modelers are needed to build the models that achieve business objectives. These geeky teammates define the layout of statistical models to predict the probability of an outcome. With their expertise, the better data is supplied to the correct model, making the prediction closer to reality.
The proper alignment of these three key dimensions enables retailers to absorb all the customer’s data through past and current behavior, no matter how they shop, and allows that data to be immediately actionable. This is especially important given that IBM Institute for Business Value found 48 percent of shoppers want on-demand, personalized promotions while online, while 44 percent want the same in store. In addition, McKinsey reports 50 percent of customers think “Integrating Stores/Online/Mobile” is where retailers need to improve the shopping experience the most.
This important use of customer intelligence transforms not only a retailer’s ability to accurately determine “who,” but also to continuously refine the understanding of “what” and “why.” By uniting customer data within online and offline sales and marketing channels, retail marketers are finally empowered to identify the repeat visitor who consistently enters through paid acquisition channels; the low average order value but high margin customer; a highly-engaged shopper who has never made a purchase, and so much more.