The current extension of the promise of “Big Data” into the field of marketing has injected accountability into a field that has historically leaned more toward art than science. “Big Data” is not a new phenomenon, although it is a relatively new term coined by University of Pennsylvania economist, Francis Diebold.
I learned in my Media Economics and Technology class that marketing can trace its origins in the field of economics. It makes sense. Marketing is, at its most basic, the practice by which entities and individuals trade goods and resources. Economics is the field that is concerned with the production and distribution of goods and resources among populations. Marketing is the field that connects corporations (producers) and people (consumers). With this in mind, the evolving practices in marketing toward harnessing big data brings it closer to its origin in a quantitative and analytical field.
To prove the point that big data is not a new phenomenon, consider the areas of operations research and management science. Academics and professionals in these fields have long applied analytical methods for decision making, optimization and efficiency. It also investigates micro-behaviors that lead to macro-level patterns.
In the research paper written by Siebers, Aickelin, Celia & Clegg entitled, “Using Intelligent Agents to Understand Management Practices and Retail Productivity,” they define operations research as a practice that is “applied to problems concerning the conduct and coordination of the operations within an organization.” Furthermore, it uses scientific models that try to make sense of the issue at hand.
For example, the textbook case of a call center. Each individual in a call center plays a role and has a work relationship with other individuals within and outside the call center– peers, managers, colleagues, customers. The behaviors of each individual (how fast they resolve an issue, how satisfied the consumer is after a call, their relationship with managers who can help them with their work, etc) can lead to trends that affect the performance of the call center as a whole. If most employees in the call center take too much time being on the phone, it leads to less customers being served at the end of the day, and has the added implication of less customers being satisfied when conversations with agents take too long before an issue is resolved. Management will then invest in training employees to better handle customer issues over the phone, or find other ways to serve customers that will lead to their satisfaction. If more customer issues are resolved and customers are happy, it leads to more revenues for the company.
Of course that is a very simple example, but one that is a classic case (taught in business schools around the nation) upon which operations research is applied. I’m particularly interested in the method used by the authors of the paper I mentioned earlier in this blog. They used “agent-based” modeling and simulation, which is about “examining complex behaviors produced by simple activities.” It is related to micro- and macro-economics, but is extended to management practices (micro-level behaviors would be things like employee actions, and macro-level behaviors would be the patterns from these employee actions that impact management decisions) instead of the economy at large.
Right now, marketers are keeping an eye on the types of data they can capture– from micro-behaviors (clicks on a site, signing up for an email promotion, purchase transactions)– to understand larger patterns that could be indicators of the health of a business. Data-driven marketing and its applications is a growing development that I’m keenly observing, in order to find ways to better serve consumers, enrich lives and grow businesses.