By Kevin Pledge
This article describes why actuaries are playing a leading role in analytics and should be regarded as the real data scientists for insurance.
Analytical systems were originally described as being able to turn data into information. This was soon extended to turning data to information and information to knowledge. One vendor took this arms-race of superlatives too far and even claimed they turned knowledge to wisdom; this is clearly nonsense. But even turning data to information and information to knowledge is a stretch. Systems can only go so far; a good analytical system will structure and make data accessible so that users (people) can make sense of it. The better the system, the better people can use it to look at the data in different ways and turn it to fit with their business perspective and understanding. A better system will also support collaboration so people can understand each other’s issues and points of view. The essence of an analytical system is about enabling people to do this; people are the important part of the process.
Tom Davenport has used the term “PhDs with personalities” and Yahoo has termed the phase Data Scientist, that makes the one-time math nerd sound quite exciting. Actuaries are the data scientists for insurance, an industry built on data and statistics. We have the core statistical training and analytical thinking skills that are so important for analytics.
Data and statistics are now seen as giving companies a critical competitive advantage in all industries. In his HBR article and various books, Tom Davenport talks about companies competing on analytics. He describes four key components:
Senior executives strongly advocate analytics and fact-based decision making;
Widespread use of descriptive statistics, predictive modeling, and complex optimization techniques;
Analytics used across multiple business functions; and
Enterprise-wide approach to analytical tools, data and process.
Missing from this list is subject matter expertise; the Subject Matter Expert, or SME (pronounced Sch-Mee) is often referred to much as you would hear flour referred to in a recipe for baking a cake. However, information is not like a baked good, and SMEs are not just another ingredient. Understanding the business is the critical factor for analytics, understanding does not come from a system, but from training and experience. The system needs to be able to respond to this; business knowledge is the driver of analytics. Not only do actuaries have the quantitative skills to be the data scientists of insurance, but our involvement in everything from pricing to financial reporting give us the business knowledge to make sense of this. This business knowledge is as important as the statistical and quant skills typically thought of when you think data scientist.
The final reason I will give why actuaries should be recognized as analytical experts is our ongoing training. Neil Raden, author of Smart Enough Systems, describes how data scientists should be trained—“like actuaries”; training on the job keeps our skills relevant and up to date. Businesses hiring these so-called data scientists usually have no problem with their statistical or quant skills, but for lack of ongoing training and relevant business skills, these analytical skills often fail to materialize into meaningful insights that can benefit the business.
The SOA is now looking at explicitly training actuaries in advanced business analytics; this will start with a seminar in 2013. Insurance companies, like other industries, are discovering that analytics should not be confined to a few specialist silos. As enterprise-wide analytics gathers momentum it is my hope and expectation that actuaries will be at the forefront of this for insurance companies, and possibly in other industries as well.
Kevin Pledge, FIA, FSA, is CEO and co-founder of Insight Decision Solutions, a company specializing in business intelligence for insurance. He can be contacted at firstname.lastname@example.org.