Big data analytics is very sophisticated, involving high-level statistical work. For that reason, the whole idea of big data is really scary to business executives. As a result, they don't work well or closely enough with data analysts and professionals (the “quants”) to get the results they need.
Tom Davenport, visiting Harvard professor and noted advocate of business analytical power, says the two sides need to start talking with one another. There's no reason why business leaders can't gain a better picture of the statistical reasoning that goes behind their analytical systems, and there's no reason why quants can't become more business savvy.
His latest book, "Keeping Up With The Quants: Your Guide to Understanding and Using Analytics," co-authored with Jinho Kim, makes a compelling case for bringing these two sides of the organization together for common cause. The future belongs to those businesses that effectively employ analytics to understand their markets, customers and operations.
Both sides have something to learn as Davenport and Kim illustrate. Business leaders, for example, should approach the relationship along the following lines:
• Give the analysts enough time and attention to ensure that they understand the problem from the business perspective, Davenport and Kim advise.
• Make available the time and attention of people within your organization “who can help analysts understand the details of the business situation.”
• Have a firm understanding of “the time and money necessary to build the solution, and jointly agree on this as a proposal.”
• Learn enough about the underlying math and statistics “to have a general idea of how the model works and when it might be invalid.”
• Don't be afraid to politely push back if you don’t understand something, and ask for a different or better explanation.
At the same time, analysts and quants need to make it easier for business decision makers through the following approaches recommended by Davenport and Kim:
• Have a good working understanding of the business overall, “and of the specific business process that the quantitative analysis will support.”
• Understand the “thinking style” of business decision makers, and the types of analyses and outputs that will influence their thinking.
• Be able to develop effective working relationships with key people in your organization.
• Use the “language of the business” to explain what benefits and improvements analytics can provide.
• Provide business decision-makers “with an accurate estimate of the time and cost to develop a model and related capabilities.”
• Be patient – if business decision makers don’t understand the project, or are skeptical of the predicted benefits, try again with different language.
• Have a structured process for eliciting the information and business rules they need to build their model.
• Help business decision-makers think about such broad aspects of the problem “as the framing of the decision involved, the stakeholders for it, and the organizational capabilities necessary to implement a new solution.”
• Develop new models and tool sets with a rapid prototyping approach, Davenport and Kim advise. “so business decision-makers will see something of substance very quickly and can provide feedback on it.”
Joe McKendrick is an author, consultant, blogger and frequent INN contributor specializing in information technology.
Readers are encouraged to respond to Joe using the “Add Your Comments” box below. He can also be reached at joe@mckendrickresearch.com.
This blog was exclusively written for Insurance Networking News. It may not be reposted or reused without permission from Insurance Networking News.
The opinions of bloggers on www.insurancenetworking.com do not necessarily reflect those of Insurance Networking News.
