- Online Exclusives
- Contact Us
I read the insurance trade magazines to stay abreast of what’s happening at the top levels of insurer management and how boardroom decisions eventually find their way to claims management on down to the street level adjuster.
Almost every issue of any insurance trade magazine contains an article about predictive analytics (PA) and how data management is a great tool for insurers to manage projected costs, actuarial models, loss ratios and fraud detection.
Just a few years back it was being spoken of as the way of the future. Today, predictive analytics has arrived.
At a recent conference, I ran into a couple of gentlemen in the process of setting up a program with estimating software creators that will incorporate predictive modeling into the estimating software, specifically targeting water losses. Their goal is to have adjusters compare their own estimates as well as the estimates provided by contractors with predictive data built into their estimating program.
If you’re in water mitigation you need to understand what this is and how to protect yourself against the reduction of your invoices by application of predictive analytics. It may not have happened yet, but it will.
On the defensive
First, let me explain how they’ll use PA against you.
They want to be able to say that, for a given water loss, under specific criteria like square footage, class of water, weather conditions, elevation, location and several other qualifiers and quantifiers, the cost should fall within a certain number.
So, for example, a water loss that affects a standard single family home affecting a 3,000 square foot area and then considering all the other qualifiers that they’ll devise, they should not pay more than $4,500 for water mitigation.
The justification for their position will be that they’ve done predictive analytics on 20,000 water losses and compared all the variable data to establish common denominators and developed averages on comparable losses. This data will tell them how much the mitigation bill should be.
The people who created these programs are statisticians. They’re selling these programs to the insurance industry and encouraging insurers to hold the contractor to the number established by PA. To their credit, they’re also telling insurers that if the PA number is higher than the contractor’s invoice, the insurer should pay the higher amount (provided the work was actually completed).
I want you to read some comments made by upper management types of insurance companies who are pushing predictive analytics.
“Fraud is one example where analytics is beneficial. Analytics can improve insurer results by:
Understand that when insurers speak of fraud, they not only refer to outright deception found in accident injuries or disability fraud but they also consider it fraud if a contractor’s invoice is higher than it should be. This is not to say they’d mount a criminal case because a vendor charged more on a water mitigation claim than what they believe should have been charged. But when they have their boardroom discussions, you can bet they talk about such scenarios as fraud and they want to resist it with PA.
When to expect it
First, in my opinion, PA will first be utilized by the same insurers who require you to use specific estimating software, limit your unit cost pricing and impose other restrictive business practices in exchange for the promise of volume.
Also, know that PA is a generic term that may be branded under different names. Basically, it’s historical data collection on specific claim criteria that they’ll try to use on you (often times after you’ve done the job).
What mitigators can do
Now, how are you going to overcome PA when imposed on you unfairly? For example, let’s say you’ve finished and billed a job and an insurer runs your numbers through their predictive analytics. You are told your bill is 20 percent too high and the insurer doesn’t want to pay.
At least if they told you beforehand that your estimate would be compared and analyzed and possibly negotiated, you’d have the option to adjust your price or walk away from the job. But, if they wait until you are finished and you’ve invested time and resources to mitigate the insurer’s exposure, you now have a collections challenge.
Without any argument, you could just lien the property — but that’s not immediately going to put money in your pocket. And it is going to sour any relationship you had or hope to have with this adjuster.
So, I recommend that you sit down with them and show them your supporting documentation and explain why your numbers were more than what their predictive analytics came up with.
It would also be fair to ask for their full PA report that analyzed the loss and your numbers. If you review it you’ll be able to compare and counter their conclusions based on data you have that supports your methodology. After all, predictive analytics involves averages and no two losses are exactly alike. In fact, no 10,000 losses are exactly alike.
It wouldn’t be a bad idea to have an understanding with them that, when your numbers fall below their PA, they should allow you an opportunity to add to your invoice. It could be you missed something. If you get a negative reaction to that suggestion, then you know that this adjuster or these insurers only want to use predictive analytics when it suits their interests.
Peter Crosa has been a licensed independent adjuster for more than 35 years, handling insurance claims throughout the United States and Latin America. Since 2000, he has traveled across the country conducting seminars and speeches on the topic of marketing restoration services to the insurance claims industry. He is author of the 2013 Restoration & Mitigation Contractor's Guide to Insurance Repair Marketing. Visit his website at www.SSHCA.net or e-mail him at Peter@SSHCA.net to ask a question or to request his free Claims Marketing Tipz.