Aon Construction Group’s data scientist on what to include in your contingency plan
by Elizabeth Manning
May 8, 2019

Your leadership team spends hundreds of hours agonizing over any one project's schedule—what workers should be hired for which phase of the project; what materials should be shipped to the site when; when and how often project owners should be brought in for a status update; and more. In a perfect world, each detail and project variable would be planned out. But then there's the world we live in, and it's not so perfect. Any number of things can go wrong in an environment where there are so many moving parts (and people). And often, many things do go wrong.

Historically, construction business owners would rely on their best guess. The strategy may be, "I know that I usually have this many delays and this many change orders, so that's what I'll plan for on this project." Insuring the uninsurable—or mitigating risk on your jobsite—is no simple task, but you can do better. Construction companies are now using data analytics to pinpoint a more accurate look at potential "uninsurables" and reduce costs and delays. Providers like Aon, a professional services firm specializing in commercial risk, are helping facilitate just that.                        

In his current role, Mike DeLio is a data scientist for Aon’s Construction Services Group. He helps lead the team’s initiatives around bringing data-driven products to the market. The team is focused on successfully bringing third-party tech into the construction space, the impacts it may have on contractors’ risk and how that transfers to insurance products. Below, learn from DeLio about what you should consider to better protect your company from the uninsurables and one of the most difficult variables to insure in our current construction climate. 

CBO: What does an uninsurable event or attribute look like?

MD: There’s a lot of variables here. We are fortunate that construction has been so steady and hasn’t changed much over the past couple years. Over time, our clients have built in contingency plans for projects. With the increased use of data, we are now able to provide them with an insurable product. A few good examples of insurables include the following: weather delays, lack of labor, political risk (such as tariffs on construction materials), labor disputes and projects that have been awarded but not started yet.

Our big focus lately has been on the weather aspect of construction insurance. The methods contractors use to plan for weather is sporadic. Some look at data—like 20-year averages of weather patterns—and plan that way. Some simply listen to the opinions of project managers on what they think and deal with on a regular basis. Our goal is to use data analytics to plan more accurately.

CBO: What measures can contractors take to protect themselves from the uninsurable?

MD: The best approach I would recommend to contractors is using what data is available to better estimate project schedules. Financial teams are doing large calculations where they relate to their financial models. A big part of that is the contingencies they have built in for delays or other uninsurables. For many contractors, a half-inch of rain on a civil project could shut them down for the entire day. Depending on the construction project, that could be $100 thousand to $200 thousand in losses. You can plan for that, though. Come to those calculations and think about what the actual costs of uninsurables could be. When you have those calculations, you can begin to put together an approach for how to plan around the uninsurables.

CBO: What is the point of no return? Do you have an example of a contractor seeing serious negative effects on its business because of uninsurables?

MD: Recently, California had that massive amount of rain after a 5-year drought. We had a client that was working on a stadium there. The rain made the client miss the scheduled opening of the stadium, which meant they had to move actual games. It was a massive claim against them. That was a great example of how bad it can be on a business. 

Pipeline contractors are a good reference, too. They will dig miles and miles of trench for a pipeline. A good rain will come through, and one of our clients had $25 million in losses because the contractor had to re-dig the entire trench. Every client deals with these weather risks.

CBO: What are some out-of-the-box aspects of an insurance plan that contractors should consider?

MD: Our recommendation is to try to use as many outside resources as you can. Pull as much data as you possibly can. You have insight into this 'mystery bucket' of contingencies—you just may not be using that insight. Break down the data and try to understand it. We’ve seen contractors hire meteorologists so they can plan around active projects and look at historical issues with weather. We’ve also seen contractors hiring data scientists who can look at general trends. So far, the data scientists are mostly focusing on the claims component and catastrophic risks, though. But I think that is changing.

CBO: Are there any new areas of protection contractors should be considering?

MD: I keep going back to the weather delays, but it’s because variations in weather have become more and more unpredictable in recent years. And this is something that almost all contractors face. However, any insurance component that covers business interruption risks is useful to consider. Every contractor needs to be aware of them and focused on gaining a better understanding of those risks.