Sun Tzu and How to Win the Data Wars

Successful data strategists adhere to four basic principles: prospect, communicate, optimize and protect.

Did a Chinese general from 500 B.C. know more about the current state of data and analytics than many modern insurers? 

Sun Tzu, the over-quoted author of The Art of War, wrote about battle – not binary code – yet his most famous saying perfectly captures the dilemma facing today’s insurers attempting to embrace data analytics: “Tactics without strategy is the noise before defeat.”

The buzz around big data is as loud as ever, but life and health insurers report mounting pressure to define a strategy and deliver a return on investment, according to RGA’s 2019 Global Life and Health Data Analytics Survey. Many are struggling. While eight out of the 10 major multinational organizations participating in RGA’s online survey reported having a data analytics strategy in place, half were in early stages of putting this plan into practice and one had not yet begun.

What accounts for the delay? Part of the problem may rest on a simple misunderstanding. It can be tempting to make the mistake of classifying data strategy solely as an information technology (IT) function, focusing on the back-end processing and management of ever more complex and voluminous data sets. But transforming a collection of initiatives into a successful data and analytics program requires more than efficiently managing ones and zeros. A true strategy should enable a carrier to maximize business value from data.

Technology is an enabler along the journey, but, to succeed, a carrier must first agree on the destination. Is the enterprise seeking to enhance in-force management, generate more leads, accelerate underwriting, engage more customers, manage claims more efficiently or something else? In RGA’s Data Analytics Survey, carriers identified the business areas most likely to be transformed by data analytics over the next three to five years. Eight of 10 participating insurers agreed that data analytics would have a “high” or “very high” influence over practices within distribution, underwriting, claims management and marketing/branding functions, with 60% of respondents sharing the intent to invest significantly in data-driven accelerated underwriting.

See also: Understanding New Generations of Data  

Next, the insurer must understand the many data sources available, whether traditional (fully underwritten) applications and financial disclosures, digital sources such as electronic medical and prescription records and insurance-linked wellness programs or more leading-edge information gathering via social disclosures and “Internet of Things” devices. Interestingly, only one carrier in RGA’s Data Analytics Survey acknowledged using wearable information such as steps, sleep and heart rate monitoring from wearable devices, but 60% of respondents plan to use wearable sources in the future. 50% indicated plans to employ “digital fingerprint” data that draw on social media disclosures, although none of the respondents use such data sources today.

To respond to these questions, successful data strategists adhere to four basic principles:

  • Prospect — Effective data valuation – the evaluation of internal and external data sources – and processes to identify, prioritize and acquire new data sources are essential.
  • Communicate — Carriers with successful data strategies have focused on understanding and communicating the data assets available within the enterprise. 
  • Optimize — Extracting maximum value from the data available generally involves identifying new or underused data assets, enriching existing sources and monetizing data or data-driven algorithms.
  • Protect — Data protection is an essential consideration. Carriers must constantly weigh how to best eliminate or mitigate risks related to data privacy and protection. 

Know the Landscape

Defining a data strategy begins with these four principles, but it doesn’t end there. As Sun Tzu noted, “He will win who knows when to fight and when not to fight.” Before pursuing a plan of attack, Tzu argued that the successful general first surveys the field of battle and evaluates the strengths and weaknesses of both armies. An effective data strategist must study the industry landscape and determine what information is available, relevant and compliant.

The data and analytics field is advancing more rapidly than regulatory rulemaking in certain markets, and carriers are challenged to plan for the future while regulatory constraints and expectations are still shifting. A “guidance map” can help, so long as it is revisited regularly. In most regions, a single piece (or two…) of legislation tends to govern overall approaches to data use. Building off this legislation, carriers can establish overall frameworks to guide risk mitigation and anticipate which data applications are likely to be acceptable. This is a start – not an end – and insurers must continue to track and respond to overall regulatory change.

Conducting a data inventory, either through manual fact-finding and in-person interviews or by purchasing an automated system to crawl available databases to catalogue them, is another important step. Aside from revealing gaps, an inventory can democratize awareness of available data beyond a small coterie of experts and help carriers draw on the collective insight of the broader organization. RGA’s Data Analytics Survey asked insurers which data sources they currently use within their organizations. In the underwriting function, the top data sources used today were claims history (60%), prescription data (50%), lab/exam and motor vehicle (40%).

Top Down or Bottom Up? 

Armed with greater insight, insurers can draw up a “target list” of data sets to pursue. This is easily the most resource-intensive task facing any data strategist, but it can also be the most rewarding. When determining the best approach to developing such a list, size doesn’t matter, but organizational maturity does. Our favorite general could have been advising any insurer’s board room when he wrote: “He will win who knows how to handle both superior and inferior forces.”

A bottom-up method is best-suited for a larger, better-resourced data and analytics team. With this approach, companies dedicate a team of seasoned professionals to systematically explore available data sets throughout an organization for untapped opportunities. This requires a deep understanding of market conditions, the capacity to methodically break down big data sets into more manageable segments and the freedom to delay immediate return on investment. The team should regularly meet with business leaders to evaluate progress.

See also: Data Prefill: Now You See It, Now You Don’t  

A top-down approach has more in common with the fail-fast ethos of a startup and works best in leaner, more limited and less structured organizations. Participants brainstorm business problems that can be solved as new and interesting data sources emerge. Rather than examining all available data, set by set, resources are focused on gathering input from business leaders, synthesizing project ideas, evaluating what business needs benefit most from a data-based solution and then coming to a consensus around concepts with the greatest chance of success. This approach generates results faster and with less investment, but, because it relies heavily on the knowledge and guidance of a few executives, it can miss opportunities. 

Tzu realized more than a millennia ago that, to win, any enterprise must out-think, rather than out-fight, an opponent. When it comes to today’s modern insurance landscape, military metaphors only extend so far. Still, it’s undeniable that a new competitive contest has emerged over data, and an effective strategy will distinguish the victors from the vanquished.

Curious? Contact us to discuss techniques to develop an effective data strategy in your organization.


Jordan Durlester

Profile picture for user JordanDurlester

Jordan Durlester

Jordan Durlester is executive director of data strategy at Reinsurance Group of America. He,builds and scales advanced analytics organizations and implements actionable data strategies designed for specific markets.

Read More