As customers strive to make data-driven business decisions, you may soon find yourself selling to chief data officers. But here's a twist: Do you need to hire data scientists within your own business?
At first glance, data scientists develop and deal with complex algorithms, work with big data systems like Hadoop, and spend their time crunching numbers to find business trends.
But take a closer look and data scientists often aren't geeks locked in a back room. As IBM puts it, "Whereas a traditional data analyst may look only at data from a single source – a CRM system, for example – a data scientist will most likely explore and examine data from multiple disparate sources. The data scientist will sift through all incoming data with the goal of discovering a previously hidden insight, which in turn can provide a competitive advantage or address a pressing business problem."
So how might a data scientist actually help an IT service provider? Here are a few potential roles:
- Upselling Established Customers: Finding patterns between your best-paying customers, how frequently they adopt next-generation services, and how comfortable they are with potential rate increases.
- Finding New Customers: Analyzing your CRM, marketing and email systems to spot potential sales leads, hidden engagements and would-be buyers that you may have otherwise overlooked.
- Firing Your Worst Customers: Pinpointing customers that call your help desk the most, pay you the least money, and complain the most about lousy services.
What's the biggest challenge facing would-be data scientists? Dan Mallinger, from Think Big's Data Science Practice, puts it this way: "Spend as much time learning analytics communication as learning models. The popularity of machine learning has led to swaths of data scientists who hunch over a computer analyzing data but can't communicate the results."
