Big Data & Analytics Heroes
Edward Kwartler, director of advanced analytics at Liberty Mutual Insurance, is this week's IBM Big Data & Analytics Hero.
What is the market still missing for big data and analytics to really deliver ROI?
I think four things need to occur to start realizing the potential of analytics.
- First, data scientists need to remember the value of parsimony. The volume of data, and improved processing power, coupled with exciting academic methodologies doesn’t give the analyst free rein to throw everything at a problem. The best analytics are more judicious and practitioners need to remember that. I like to remind my team about big enough data.
- Second, analytics professionals need to value and incorporate qualitative inputs into their work, including understanding the business problem. Don’t go fishing for an answer, have a clear objective and learn from the business professionals affected.
- Next, IT and analytics departments need to improve the data integrity within organizations. Often analysts spend half their time doing nothing more than organizing and vetting data before really doing analytics. The two parties must collaborate on efficient and accurate data marts.
- Lastly, the current generation of senior stakeholders needs to become more data fluent. Many of the long standing corporate powers realized success with qualitative observations and experience. Data driven decisions can be uncomfortable for these stakeholders so learning a bit about analytics can supplement and improve the dialogue.
Should tomorrow's generation acquire analytic skills no matter the degree? Why or why not?
I think that tomorrow’s generation of professionals has to learn some basic analytics. In a world driven by computers and becoming more automated, having a basic knowledge of analytics and even the logic of coding helps make sense of these changes. Without it the world will become a series of black boxes in which directives are accepted at face value. For the non-analytical professional that could be a difficult world to thrive in. However, I am confident we will always value the qualitative and creative process. In fact, I think the best analytics practices marry art and science. Remember that numerical imagination leads to efficient means but it is narrative imagination that justify the ends.
How can big data combined with analytics improve the world we live in in 5–10 years?
The three most promising analytics and big data initiatives of the next five years are centered on speech analytics, text mining and spatial data. Data fluent organizations are waking up to the fact that there is a lot of unstructured natural language and also organizations have the ability to understand where the data itself is being created. Natural language can be in the form of customer call recordings or notes in customer accounts. Spatial information is often collected based on cell phone locations during app use. Understanding these new data steams will result in algorithm lift and new insights into customer behavior.
View all of our Big Data & Analytics Heroes here on the Hub, and look out for next week's Hero: William McKnight, president of McKnight Consulting Group.