Tackling Data Analytics Means Knowing What Questions to Ask

More than ever, insurers are sitting on treasure troves of data, just waiting to be extracted from policy administration systems, claims files, insurance application information and customer service center engagements, and analyzed.


More than ever, insurers are sitting on treasure troves of data, just waiting to be extracted from policy administration systems, claims files, insurance application information and customer service center engagements, and analyzed.

What is needed is someone who understands how to put all the pieces together in order to turn piles of data into insights. The good news is that you don't have to go back to school and become a “data scientist” to make this happen. What is needed is common sense along with a good idea of what the business wants.

That's the gist of "Lean Analytics," a book that strips the art and science of data analytics down to its bare essentials so that any business manager with a passion for better understanding the world around him or her can get in the game. While the authors, Ben Yoskovitz and Alistair Croll, wrote the book for the startup crowd, there are lots of valuable nuggets of advice that can help a manager within an organization of any size or legacy get started on the data analytics journey. Let's face it, any data analytics effort is a “startup” in and of itself.

Yoskovitz and Croll's work was reviewed by “lean startup” guru Eric Reis, who explored their advice in some detail:

1. Start with metrics in mind: Typical metrics include churn, customer lifetime value, viral coefficient, acquisition cost, uptime, and engagement. For example, an example of starting with a metrics-oriented goal may consist of reducing churn down to 5 percent per month. “Knowing what normal looks like is essential,” Yoskovitz and Croll advise. “If you don't know what normal is, you can't tell if your efforts are paying off. You don't know if you're at a point of diminishing returns and should focus on something else.”

2. Find the one metric that matters: As you start off with metrics, select the one most important metric to the business, and focus on it like a laser beam.

3. Recognize that not everyone will be happy with the results: Yoskovitz and Croll point out that the goal of data analytics isn't to try to solve problems for “safety and predictability.” The goal is “to take risks, and to uncover the non-obvious and the unpredictable.”

4. Ask good questions: "Your customers leave a trail of digital breadcrumbs with every click, tweet, vote, like, share, check-in, and purchase, from the first time they hear about you until the day they leave you forever, whether they’re online or off,” say Yoskovitz and Croll. “If you know how to collect those breadcrumbs, you have unprecedented insight into their needs, their quirks, and their lives.

This insight is forever changing what it means to be a business leader. Today’s leader doesn’t have all the answers. Instead, today’s leader knows what questions to ask. Go and ask good questions."

Joe McKendrick is an author, consultant, blogger specializing in information technology. This blog originated on Insurance Networking News, a SourceMedia publication.