March 18, 2018

Dax Craig: Developing an Overarching Predictive Analytics Strategy in Work Comp

By Dax Craig, CEO, Valen AnalyticsDax Craig

As the work comp market continues to become more segmented and complex, carriers are increasingly turning to predictive analytics in underwriting to stay competitive. The fact that the historically troubled line of work comp is now profitable for the first time since 2006 at the same time we see market share growth for companies known for their holistic use of analytics, like Travelers and Berkshire Hathaway, should act as a wake up call that the competitive landscape in work comp is becoming more sophisticated.

Dax Craig Valen LS WC 2015 Market share

In fact, Towers Watson’s 2015 Predictive Modeling Survey found that the use of predictive modeling in risk selection and pricing for work comp increased 47% to 61% from 2013 to 2014. While this shows that insurers are moving in the right direction, there are too often misconceptions about how to create a real predictive analytics strategy versus simply adopting predictive modeling. It’s still a new area for many, as Valen’s 2015 Summit Survey showed: a majority (56%) of insurers have been using analytics in underwriting for less than two years. Most insurers still invest in one-off siloed projects that only improve a particular segment of their business without a plan to align it with the overall business goals of the company. The defining factor that separates a project and a strategy are tied to how well an insurer implements predictive modeling.

Implementing predictive analytics in an organization is a significant undertaking, and finding the best predictive model is only one part of a much larger process. Everyone in the organization must be on board and in sync with the project rollout. No insurer will handle this the exact same way, but there are three key components that will ensure you’re on the right path to a successful predictive analytics strategy.

Secure Executive Sponsorship
It begins with winning over the decision-makers, and that means doing your research and being prepared to field the tough questions. Some questions you need to answer include:

  • How will we prove that a predictive model will produce results? What is our proof of concept?
  • What are the agreed-upon metrics to measure success? (Loss ratio, price competitiveness, premium growth?)
  • What management reporting will you put in place? If it can be measured, it can be managed.
  • How do we know if a predictive model is giving us new insights vs. telling us what we already know?
  • What is our risk appetite for this initiative? What are the assumptions and sensitivities in our model and how will those impact projected results?
  • What is the plan to integrate the model within our existing workflow?

Once you have answered these questions and secured C-suite approval, keep in mind that there should also be a leader assigned to handle the overarching implementation. If you don’t have an executive owner for innovation and analytics – it’s tough to execute a solid strategy. Once this is in place, it’s time to gain adoption from the rest of the organization.

Aligning the Front Lines of Your Team
One of the most pervasive reasons for insurers’ lack of strategy is the cultural hurdle that comes from not having organizational buy-in from employees. This is deadly for efficiency. According to global consulting firm McKinsey, “Successful adoption requires employees to accept and trust the tools, understand how they work, and use them consistently. That is why managing the adoption phase well is critical to achieving optimal analytics impact.” Essentially, you can do everything right, but if the people on the front lines fail to utilize predictive analytics, it will all be for naught – and executives understand this. According to Valen’s Summit Survey, 82% of participating P/C executives cite underwriter adoption as a big concern when deciding to implement predictive analytics.

To put their minds at ease, follow a thoughtful, straightforward process that involves all the stakeholders early and often. Make the business case for why implementing predictive analytics makes sense for your organization, seek out the appropriate influencers who can help communicate and provide feedback, and create transparency by showing what data went into the model and why, to help underwriters specifically understand why it’s an efficient tool. Once you have buy-in, it’s important to implement a structured training program, provide support for questions and issues that may arise, as well as enable a test and learn culture, allowing a few months to iron out the kinks.

Ensure Successful Implementation
Once you have secured buy-in at both C-level and the rest of the staff, resources and capabilities should be assessed to ensure successful implementation. That means data considerations like ensuring a good sample size, accounting for selection bias and addressing blind spots are key. Modeling best practices must also be vetted, such as establishing a process to standardize and normalize data, identifying how the model will be used (for either automated use cases or results that will have more of a human element), and solidifying a time-to-market. If it takes 18-24 months to deploy, the value of the project goes down considerably.

Implementing a predictive analytics strategy starts by first identifying a strategic business priority. Once insurers are aware of their end goal, it will be significantly easier to garner the support and buy-in needed from top leadership as well as the rest of the company. Make sure the leader of this initiative is ready to take on the responsibility of guiding the strategy as it may shift to each company’s needs and that the strategy itself is always quantifiable and can be tested. Having objective criteria to evaluate different options will help ensure the right decisions are made, and arm the organization with the necessary data to justify the investment down the road. By sticking to these three key areas, work comp insurers will be able to create a sound and sustainable strategy and optimize their success in using predictive analytics.

About Dax Craig
Dax Craig is the co-founder, president, and chief executive officer of Valen Analytics. Based in Denver, Colorado, Valen is an advanced data and analytics provider for the property and casualty insurance industry. The company leverages its large consortium data assets to help carriers price insurance policies more accurately and achieve lower loss ratios.

Prior to founding Valen in 2004, Dax was founder and CEO of Xertex Technologies, which was acquired by global leader in the wireless antenna industry, Centurion Wireless. Dax proceeded to serve as vice president of global business development at Centurion, where he was directly responsible for global business development including sales, market definition, market segmentation, market research, strategic planning, and market development.

Dax graduated from the University of Tulsa with a bachelor’s degree in business administration and marketing. He earned his MBA in finance from the University of Colorado at Boulder.

About Valen Analytics
valen analyticsValen Analytics is an advanced data and analytics provider for property and casualty insurance companies. We work with insurers who are actively looking to improve underwriting profits by driving growth and/or lowering their loss ratio. Our customers are focused on increasing competitive pressures, fighting adverse selection with innovative solutions, and raising awareness for the impending “experience gap” in underwriting with initiatives such as Tomorrow’s Talent Challenge. Our customers span many lines of business including Homeowners, Workers’ Compensation, Commercial Auto and Telematics, Commercial Package, Commercial Property, and BOP. Learn more about Valen at

  • RSS
  • Twitter
  • LinkedIn