Leveraging the Power of Data in Insurance with Harry Storck from Protective Insurance

James BenhamOn this episode of the InsurTech Geek podcast, talking about leveraging the power of data and insurance with Harry Stork from Protective Insurance. The InsurTech Geek podcast powered by JBKnowledge is all about technology that's transforming and disrupting the insurance world. We'll be interviewing guests and doing deep dives into specific tech we see changing the industry. We're taking you on a journey through insurance tech, so enjoy the ride and geek out. And we are back with another episode. With me, my illustrious cohost, Rob Galbraith. Rob, how's it going?

It's going well, James. Excited to be here today, and, it is definitely getting hot down here in Texas. It is. It's summertime. School's ending. School's out fall summer. It totally is. This is the last week of it and, you know, it's gonna be a hot, hot, sweaty one in Texas, but that's okay. With us today, Harry Stork from Protective Insurance. Harry, how's it going? Everything's going well. Not too hot here in the Baltimore area, a little cloudy, but,

Harry Storckspring is in the air. Yeah, it is. You

James Benhamguys are usually about a month or two behind us, but enjoy that springtime weather while you got it. Go eat some crabs. Did I tell you my funny story last time we saw each other about crabs in Baltimore?

Harry StorckI don't recall, though you may have, but I'll I'm have to hear it

James Benhamfrom Louisiana. That's where I'm born and raised. I was visiting some good friends and business colleagues in Baltimore, and they said, we're going take you to the local place, the local Baltimore crab place. We did, and it was great. It was delicious. The waitress came by and I asked, I said, hey, by the way, where are these crabs from? She goes, oh, Louisiana. Jimmy went all and the way to Baltimore to eat my own Louisiana crabs. So that was my funny Baltimore experience. I love it up there.

Nice place. And we're glad to have you on today. Harry, we're going to talk about loss prevention, we're talking about innovation and technology and insurance, we're going to talk about all kinds of things. Most important topic is about data. Before we do that, I want to talk about you and just introduce guests to you. You went to Loyola University in Maryland, got a business degree, went to University of Michigan, place my daughter wants to go to school by the way, she wants to get a musical theater degree from UMich.

That should kill you, Rob. Sorry about that. It does. It does. And then you went to Darden for some executive leadership curriculum, but you've gotten some good executive leadership, you got a good business degree from a great school, and then you jumped in to a few different jobs. It looks like early on, you were actually

Harry Storckat waste management, is that correct? Correct. Yeah, I was there for about eighteen years in a variety of different roles up and down the East Coast. So kind of caught my teeth in safety and in operations in the solid waste business. Yeah, in a very dangerous business.

James BenhamI'll tell you that I was a city councilman here in College Station for two terms and we ran our own solid waste collection program And it was by far the riskiest, highest frequency, highest severity, highest risk operation we had was all those, rolling tanks.

Harry StorckYeah, it's, you know, when big vehicles are moving and humans are involved, it can be a, a dangerous combination. So, safety first as they say. Yeah. Absolutely.

James BenhamWell, I mean, let's let's go back to when you were a kid, though. Did you grow up in Maryland? I did. I grew up outside of the DC area in Bethesda, Maryland, just outside of Washington. Nice. And so they always like asking people like when you're growing up in Bethesda, know, what was that? What was your career dream when you were a kid? Like what was the big idea back then before you went to college and got real about the job? Yeah, I would think, you know, probably like many youth being a professional athlete was

Harry Storckfront of mind for many in their younger years. I didn't really have a clear career path until I kind of made my way into my studies at Loyola. My father was very successful as an executive and a sales leader in an organization. So the early part of my career, I was in sales and operations at waste management and then made a shift over to kind of more of an executive leadership role, when I moved into the vehicle leasing world with PHH R. Vow, which is now, Element Fleet, the largest fleet management company.

And after I left there, of got into vehicle safety and then into the insurance space. So I'm kind of, getting into the insurance world, you know, kind of the last, eight years of my career versus where I started. Yeah, that makes sense. You

James Benhamcertainly understand the necessity for insurance and for mitigating risk. I mean, you saw it firsthand with some pretty hairy operations there. Yeah, the human

Harry Storckanimal is in control of their consequences or their behaviors. We're creatures of habit, shifting driver behavior, using technology to do that, capturing the data that can help you understand and identify and correct.

James BenhamYou know, those are things that have been around for a while, but doing it day in and day out is much, more challenging than the theoretical angle of it. Now, what did, data capture look like back in the day though? I mean, when you were trying to do this in waste management or almost two decades you were there, this was way before, I mean, started right when the internet commercialized and way before mobile apps and everything else. What did data collection look like then? Yeah, well, in my early sales

Harry Storckcareer, you know, my data capture was the three by five index cards that you collected all your contract expiration dates on your prospects that you were calling on and kept a manual tickler file of who are the people that you should be engaging with at the right time, back, in solid waste business. It's highly, dependent upon service agreements that are in force similar to the uniform rental business and pest control, things of that nature, the recurring monthly revenue type of a business model. So having agreements that were valid and in force, was the big, big data issue and everyone knew it was important, but no one really did anything about it.

A lot of the growth in the solid waste business was based on acquiring family businesses and their record keeping at that level was mediocre at best. So the biggest data capture, early on in my career or exposure to that was really in the operations part of the business, measuring productivity,

James Benhametc. Gotcha, makes sense. Yeah, productivity, and I've always felt like productivity analysis, productivity tracking and safety and lean all go hand in hand. You know, like they're cousins of disciplines, right? Because ideally when you're making somebody more productive, you're eliminating steps that can also eliminate a lot of risk, right? Absolutely.

Harry StorckSo, you know, risk is a function of miles driven or hours on the road, in the auto space. So whatever you can do to minimize that exposure, through, operational shifts, where possible, you know, that that has a positive outcome on your risk profile.

James BenhamSure. Yeah, we kind of felt the same way with construction sites and did a whole lot of years of analysis of workers walking across job sites. The simple fact is if they walk less across the job site, there's a lot less risk, less risk to get run over or hit something or vehicle is going to collide into them. And so you kind of have to look at it like, like close cousins there. Tell us about your current role. I believe you're vice president of safety services at Protective Insurance, which is now part of Progressive, right?

Right. So, Protective,

Harry Storckhas been over its life a niche player in commercial auto, specifically in trucking, and then also offer workers compensation coverage vertical and a couple of other, spaces, but primarily we are, in the commercial auto space. So Protective was in business for over ninety years based in Carmel, Indiana, very well recognized as a niche player in that space. And we were acquired two years ago by Progressive to be really truly a complementary tuck in acquisition where the fleets that we ensure in the heavy duty regulated space are 25 power units and above and protective strong suit has really been more in that, you know, three to five vehicle space and mostly light duty vehicles in what they call their version of trucking was maybe more defined with pickup trucks and other service vehicles where we are truly in the tractor trailer business.

We also do supporting FedEx, Amazon, few other of what we call our program business. So my role at Protective is I lead our loss prevention team. We have a team of fleet safety experts and workers compensation experts that are geographically dispersed across the country. My reps, I'm very, it's a double edged sword. My average, tenure is about thirty years in truck safety. So, highly expert and tenured team that's very well respected in the marketplace. They work in an assigned portfolio business model where each one of my, field reps has got about 40 clients that they support.

Truly. I guess you could say that our two customers that we work with are our policyholders, to help them be safer and more compliant, which also benefits our second customer, which is our underwriting colleagues that are involved in pricing the risk transfer.

Rob GalbraithCool, thanks for that, Rob. Harry, I'd love to dive in a little bit more. You mentioned kind of your reps in the portfolio and the two different customer that they serve, right? The client and then the underwriters. You know, think when I think about loss prevention and I think about data capture, particularly when you're talking about vehicles, I think there's always been a bit of a skepticism sometimes on the part of the insurers of, you know, what is this data for? How are you gonna use it?

And the very real concern that it be used in the underwriting process and ultimately trickle down to pricing. So, there's always been some reticence in many sectors for sharing that data. So, maybe you can just talk a little bit about, are you able to create kind of win win scenarios? What are those conversations like? And has it evolved over time in your experience? Yeah, that's a very, very accurate description,

Harry StorckRob. So traditionally insurance has been priced using proxy, type data. You know, if you are if you fit into this bucket and this bucket and this bucket, the crystal ball says this is your price. Commercial auto over the last ten years or more has been a very unprofitable business. And so what is the number one rule in running a business? It would be know what your costs are and charge more and then you make a profit. Okay, so that's very, very difficult in insurance because insurance does have that necessary component of a crystal ball.

And the keepers of that crystal ball are the actuary folks that do their best to quote unquote get it right. The common levers that have been reviewed in the past have been proxy data. How old are you allowing drivers to drive your trucks? You know, what kind of mileage do you drive? What assets do you have? What freight do you haul over the road? Regional? Are you employee drivers? Are you owner operators or independent contractors? And the most robust data historically that has been available has been, years worth of claims data.

So typically an insurance carrier is going to look back three to five years at your claims experience. And that's a major contributing factor in your pricing. In our industry, another huge lever that is included is now the data that is available through the Federal Motor Carrier Safety Administration, which is, administration under the Department of Transportation, trucking is highly regulated and the Federal Motor Carrier Safety Administration, FMCSA has created universal safety scoring metrics and formulas based on peer group is kind of the foundation of it. So we have data that we'd, get through a third party source that provides us access to their federal scoring of how the United States government, reviews their safety.

It's not a perfect model, but much better than it's been historically. So matching that up with claims data have been the two primary levers. And then the last thing Rob you mentioned was about, people sharing their data, you know, telematics data and other types of vehicle technology has been around for a long time in the trucking space. Telematics, you know, they were trucking was the early adopters of telematics, but nobody wants to share their data. And the reason for that is in the old days when the company adopted a telematics platform, many would say that there was fear of big brother.

The driver didn't like their employer to know everything about what they were doing. That is pretty much gone by the wayside now. The barriers to that or the trepidation around that has subsided to a large degree because it is so commonplace now. But the new big brother Rob is now the insurance carrier that wants the data from the fleet safety team. So it's shifted up one level from driver to employer. Now it's from policy holder to insurance carrier. And over the last three to five years, commercial auto insurance rates have been going up every year by over double digit increases.

So a fleet that just was, receiving of a 15% rate adjustment one year to the next or higher, they're likely going to be a little cautious about sharing data that could possibly make that increase be even more. So we are working on a couple of initiatives now, one in particular, where we are kind of proposing a value add trade to offer a new solution to our clients to help them mitigate risk associated with driver fatigue. It is a technology that is connected through an API with their electronic logging device that is mandated by law, an ELD device that measures a driver's work schedule, hours per day, hours per week, etc.

And from that schedule of data and the ELD is a common denominator that 99% of our fleets use today. So we recognize that as a good data source, But instead of saying, can we have your data? We partnered with a fatigue science organization and said, we will provide you a new tool, a new arrow layer quiver to help reduce risk associated with fatigue. And for that value of this new product, we get access to your ELD data and we're doing that in a voluntary, way where clients are given the opportunity to opt in.

We started with a small handful of clients that we've insured for many, many years just because we know them best and they've traditionally been our guinea pigs for other new initiatives that we've started. And now we're, we're making it available to all of our companies that we ensure that come to us through third party broker agents. So we've got, we're nearing 10,000 power units on that initiative now. And we're using that data to learn more and more about how our fleets truly operate versus what they're representing to us in their applications.

Rob GalbraithSo Harry, I think that's, really fascinating. I love the kind of behavioral science element and the forensics fatigue is obviously a huge issue. I'm kind of curious when you approach it from a bit of a win win, right? You're giving them an extra tool in the toolbox kind of in exchange for the data. And I love how you're partnering with kind of, you know, trusted customers, right, that you've known and and really pilot through them first before kind of expanding more broadly throughout your portfolio and and using that really as a as a marketing tool as well.

So maybe you can share, you know, what are some other concerns that fleet safety managers have that you're either seeing from a underwriting and claims perspective or that you're hearing from your customers as it relates to fleet safety? Yeah, there's two big ones that come to mind.

Harry StorckNumber one is there is a huge shortage of CDL drivers in The United States estimated to be somewhere in the neighborhood of 80,000. So there are not enough drivers to drive the freight that needs to move today. Back when there was talk about the rail strike, the railroad strike a few months back, it was estimated that number would jump to 400,000 if the rails were shut down. So obviously not a sustainable model and all the supply chain issues resulting from that would be catastrophic. So number one is shortage of drivers and included in that topic are the vast majority of truck drivers in The United States are over 50 years old.

So not only are there not enough drivers, but the drivers that are there are going to be closing in on retirement. It's a kind of a baby boomer industry if you will, or, whatever comes after baby boomer. So that's a critical, critical thing. So we're even working with partners that can help our clients or policyholders do pre employment screening of potential driver candidates, not to ensure only that they can drive and they have a good driver's license, valid license, etcetera, But also they have a good psychological makeup to be a loyal employee over time because the cost of turning over employees is astronomical.

So driver shortage number one, a very close second and they could maybe even be interchangeable would be the power of the plaintiff bar in The United States. The plaintiff bar and maybe you've heard of the term nuclear verdicts that are out there. Whenever you're watching TV, you see all the lawyer commercials about if you're hit by a truck, give me a call and it's like winning the lottery. They are highly organized, highly professional. The least of our worries are the billboard or the TV advertising lawyer.

There are now national firms that are highly sophisticated that, maybe in the old days may be considered to be ambulance chasers. Well, these folks are not that they are highly organized and frankly way, better prepared in their legal pursuits than the defense bar is, and the defense bar even acknowledges that. So the days of a $1,500 fender bender when a truck taps into somebody at a stoplight, that loss is now a $500,000 bodily injury claim. So those are the, keeping enough drivers on the road and making sure they're safe is a key need in the marketplace.

But when you're driving the exposure or the liability of people coming after you to sue you and ultimately us fulfilling a promise of paying those claims, that has made it very challenging to have a combined ratio less than a 100. Yeah, I mean, you're absolutely right in terms of the profitability segment and the pressures

Rob Galbraithon premiums, the double digit increases, it's been a very, very challenging. I mean, you look across really any carrier in this space, how challenging it is. I'm curious, Harry, have you experienced or seen or have any thoughts on some of the technology? You talked about the nuclear verdicts that are cameras in the dashboard, whether they be external facing, internal facing, sometimes both, maybe the connected vehicles with an app, right? That's on the phone logging data somehow, just to kind of look at causation, whose fault was this, etcetera, anything kind of around those spaces?

Harry StorckAll of those things, we track all of those. So every single company that we insure, we know if they have a telematics device, an ELD device, a forward facing camera, a driver facing camera, lane change technology, do they have a distracted driving app on their phone that prohibits phone use? We know if they're using collision mitigation technology that will either notify the driver of an imminent rear end collision or even, press the brakes, as well. So we track that data on all of our clients, not only do they have it, but who do they have it with?

At what percentage of their fleet do they have it deployed? And then an opinion from our loss prevention representatives, of a green, yellow, red, how well are they using the technology to get the ROI? So we have that data. One of the big steps that we took a few years ago when I took over the loss prevention team, when we used to do work for our clients like many, many other insurance carriers, and I worked for a big one before, Protective, You know, the typical loss prevention exercise was to do something right before renewal or closely after renewal saying, wow, we've got another twelve months with this risk.

Let's work with them, give them recommendations on how to be better. And we would go through a client exercise typically driven by premium size and renewal date. And we would engage with a client and get a feel for what they're doing and their safety practices from A to Z, store all that information on a Word document, convert it to a PDF file and store it in our library of all of our risk surveys. So if anybody wanted to know something about a client, you would have to go to that survey document and read it and understand what's going on.

So we said, hey, that's not good. You know, unstructured data in a narrative or comments field is not good. And we went through a very, disciplined process with data governance to say we're platform that we use to store all this data and convert everything that was previously a narrative box into hundreds of structured data fields that are numeric, they're calendar driven, they're percentages, etcetera. So now we have two years worth of this data to say, we not only know who our clients are, what they pay us and what their policy structure is, But we also have that opinion of our risk consultant to say, here's what we feel they're doing.

Are they trending to the good or to the bad? So we can give our underwriters good info, but more importantly, give our clients resources to improve whatever their current state is. So we have been using this new data platform now for two years, we just recently got it migrated over to the cloud and now we're starting a process of integrating our loss prevention data along with our claims data and the federal government scores that I mentioned earlier to now have kind of three sixty degree view of the risk and not just a very siloed view.

So that's starting to pay big dividends for us now. And the biggest reason is, we have a lot of clients that are performing really, really well. And so we don't really maybe need to be spending as much time with them, even though they're paying a good premium. Let's be more targeted and allocate resources based on gap analysis on clients that maybe aren't performing as well as they could be. And they're interested in working with my team to get better because it takes two to tango. You know, there's some people that look at loss prevention as necessary evil, but for protective, it is a cornerstone of our value proposition.

James BenhamYeah, let's talk about the future of insurance. Like let's say five to ten years from now, we're seeing rapid advancement in artificial intelligence. I mean, really rapid, like exponential advancements and the capabilities of large language models, LLMs, machine learning algorithms. And we now have inexpensive data collection devices, and terabytes of terabytes of data on all of our risks. It's really not nearly as much of a guessing game anymore. I think that's really where we're aiming at is narrowing the gap of guessing and rating risk individually based on what they're actively doing that could be more or less risky.

Do you think, and I'm just kind of positing a question about auto insurance in general, this would include personal lines auto. Do you think that there's going to be a future in the next few years where you won't even be able to buy it if you don't consent to being tracked? Well, that's definitely the case in commercial.

Harry StorckIf you're not so much being tracked, but if your performance is subpar and everyone is losing money and shoring these bad fleets, you can't be in trucking unless you have $750,000 of coverage on your policy. We offer a minimum of $1,000,000 So it would potentially put a business, out of business if you can't get coverage In the consumer space, UBI has been met, depending on who you speak with, it's either the best thing in the world or it's not so good. Early on companies like Progressive and others offered an automatic discount if you opted in on UBI to get the snapshot tool or whatever else the devices that goes into the computer in the car.

What was overwhelmingly found by many was that the consumers hated that. They didn't get the discount they thought they were gonna get. And they had a device in their vehicle that kept beeping like an alarm clock every time they went on behavior that wasn't acceptable to the device. And the overwhelming response from the marketplace was dissatisfaction. So they're kind of rethinking strategies there to say, we're not gonna give you a discount. We might give you a one time know, a parent to join, but your policy will be priced truly based on your usage and your behavior UBI that is not in place in commercial insurance except for very limited, examples.

The technology and everything that is potential, it really is a function of will the policyholder share and then you're mentioned James, they might be required to share, but then moreover, does the insurance carrier have the capacity to store all this data? Storing terabytes of terabytes of data is expensive. So it's not just some free thing that you get all this data. Can you legitimately store it? And if you can, how do you make heads or tails of it? Those, there's a lot of talk about everybody doing that, but we're not seeing it yet.

You know, people are tippy toeing into this, which you have to, but nobody is leading the way in the commercial space, at least on getting all this data. It's just not happening. It's talking about happening, but it is not happening yet. The other thing that's not happening anytime soon is the big big push on electric vehicles. Electric vehicles are not gonna take over trucking anytime soon, or even the general auto population until a major overhaul of our infrastructure is undertaken. And I'll give you an example.

A very large trucking company wanted to open a new terminal in a Midwestern town run 30 to 50 trucks out of there. They went through the permitting process with the local city council town council laid out their plans for being an innovative green company to run an electric fleet, and the entire thing was shot down because those 30 vehicles would draw more electricity than the entire town. It sounds good, but it's not happening. And the last one I would add is the autonomous truck. You know, in very limited circumstances, there are some uses of autonomous trucks or even truck patuning.

But if you get information through Atri or other,

James Benhamyou know, think tanks on the trucking space, you know, it's fifty years down the road, not ten years. Yeah, it turns out it's a little more complicated to move tons and tons and tons of cargo on a public highway than expected. I remember when the CEO of General Motors said that they would have fully self driving automobiles by 2018.

Harry StorckWe don't have room for any more bridges to nowhere. Like we've heard about out in California. There's not money to pay for that stuff. Exactly.

James BenhamExactly.

Rob GalbraithWell, Rob, bring us home. Harry, think it's really fascinating. And I love what you're talking about in terms of write this data. It can be a treasure trove, but it can also be an albatross. If you don't know how to manage it, you don't know how to make sense of it. So any last thoughts just in terms of like partnering, know there's startups out there and others that are offering help both with managing the data, collecting the data, making sense of generating risk scores, both in the commercial auto space and the work comp space.

So any thoughts about kind of Intertek startups, the ability of a traditional company like Protective to partner, how much have you done? How much have you done in house? Just kind of curious your thoughts on that. Yeah, no, that's a great topic because everyone is really looking for that answer and trying to figure that out.

Harry StorckSo, I guess you could kind of liken it to the damned if you do damned if you don't. So everyone in, the casualty lines, commercial auto in particular is very, very focused on a combined ratio less than 100. Okay, and for those that have achieved that, which we have, and we've been profitable for the last few years, even prior to progressive, took a lot of hard work and a lot of tough conversations with getting rate resources and financial resources are very, very sparse. And the key question is how much can we afford to add to the expense line with the hope of getting the ROI?

We have clients that we've insured for fifteen years and they've had telematics for all fifteen years, but they're not all profitable. So just having the technology, I make the analogy of, I got to lose weight, so I'm going to join the gym. I'm going to make a financial investment to join the gym or to buy a telematics device for my truck. But if I don't use the data or I never go to the gym, I'm not going to lose any weight. So just buying the solution versus taking full advantage of it and using it properly and understanding what human resources required to document everything you're doing so you don't make it easier for the plaintiff bar.

That is very, very challenging. And our example of partnering with a fatigue management company was really a great first step for us because it had the common denominator of the ELD device that all of our clients have, but the solution did not require any additional hardware. So they could leverage the technology already in use in their vehicle and really just use an API to transfer data and have the secret sauce take action. And so hopefully there'll be more opportunities like that or others. But as I said earlier, there's a lot of talk about sharing data.

A lot of the small companies would love to partner with an insurance company, but when you figure out what the cost is, when you multiply it out times 100,000 vehicles at $10 a month times twelve months a year, that's a huge investment. And I don't think we're at a profitability level yet that we can do that yet, or we have to pick our battles would be my answer. Okay, great answer. Well, look, it's been a good discussion.

James BenhamExcited to see the future of, telematics, the future of big data, future of artificial intelligence, all of those things are actually applying. One of your more interesting points was that real world example, you have clients who've been using telematics all along, and they're still not profitable. There still is people in process. You can't implement technology in isolation and expect it to solve or be a panacea. You still have to work on people, you still have to do training, you still have to improve your process. And even then, because it's the world, bad things still happen.

So it's still not The

Harry Storckkey thing about telematics oftentimes is the people that buy the solution at the C level that are, connected with the software team, they are not the ones ultimately that have to use the solution. And that is that frontline safety supervisor, somebody that has 20 drivers they are responsible for. They're already working eleven hours a day for a trucking company and somebody in the Ivory Tower bought a new shiny object that just added another hour to their day that they don't have to begin with. So my team helps them streamline some of that data volume to really focus on the things that are most critical.

And if your truck goes over the speed limit by two miles an hour, I don't need to be alerted by that and get 50 of those per driver per day. I just can't do anything with it. But that's what happens day in and day out. Totally understand.

James BenhamThat makes sense. Well, Harry, thank you so much for joining us. Rob, as always appreciate you co hosting brother. Absolutely great conversation. Thanks Harry. Appreciate your insights. Pleasure

Harry Storckto join. Feel free to reach out for any other areas where I can help. James, look forward to seeing you at some future events.

James BenhamYeah, I'll be at the Reuters event in Chicago in June and Reuters again Austin in September. So those are my next two events coming up. Obviously, you and I saw Austin. Yeah, I think the claims event in Austin, I'll be there. Yeah, I will be there. It's in both Rob and I's backyard. Rob, maybe we can drag you over there too. So, be awesome. And, again, to all our listeners, thank you for tuning in today to geek out for our interview with Harry Stork from Protective Insurance.

See you next time. The InsurTech Geek podcast powered by Jibby Knowledge, jibbyknowledge.com. It's all about technology that's transforming and disrupting the insurance world. I've been your host, James Benham. That's jamesbenham.com with co host, Rob Galbraith at endofinsurance.com. And thank you for joining us today. Look forward to talking with you soon. We're taking you on a journey through insurance tech. So enjoy the ride and geek out.