DIGIN 2020: Innovating for the future of auto insurance with telematics

This DIGIN 2020 webinar featuring Arity, among others, explores telematics for auto insurance and additional innovative developments in the industry.

Louisa Harbage-Edell:

“Telematics will revolutionize the insurance industry.” How many of you have heard this bold statement, and how many of you are still waiting for that meaningful change to really come to fruition five, 10, or even 15 years later? My name is Louisa Harbage-Edell, and today I’m joined by four industry experts for this bootcamp on innovating the future of auto insurance. We’ve reached a point where a large number of companies have a telematics program in place, but many still struggle to gain critical mass. In fact, less than 10% of policies enforced today across the industry are usage-based auto insurance ones. So what’s holding us back, and how can we get past these barriers to achieve the opportunities we see for telematics in the future? In today’s informal discussion, my panelists will address these issues and more. We’ll also be asking you, our audience, some poll questions to get your thoughts, and we’ll open it up for your questions via live Q&A towards the end of the session. That said, you’ll see that Q&A panel is open throughout, so feel free to drop in your questions at any time. By the end of this conversation, you’ll have ideas for how to overcome some of the key challenges still facing companies today in order to take your program to the next level, as well as key insight into the exciting new ways telematics is evolving for the future. With that, let me introduce our panelists.

First up, Gina Minick is product director for Insurance Solutions at Arity. She has extensive experience in product management as well as, prior to Arity, being at Allstate, where she played an integral role in the countrywide rollout of Drivewise, Allstate’s telematics program. She’s also served in numerous other roles at Allstate and Progressive, including product strategy and implementation, risk management, catastrophe management, competitive intelligence and marketing. Larry Thursby is a P&C insurance consultant and a former Nationwide executive. He worked at Nationwide for 25 years before retiring in 2017. His final 10 years at the company were as vice president of personalized product and pricing, and it was of course during this time that Nationwide introduced their first UBI program for consumers. After initial success in that pilot state, that offering grew into the countrywide program including OBD-II [On-Board Diagnostic II] and smartphone connections as well as connected car solutions.

Teresa Scharn is the current VP for personal lines product development at Nationwide and has worked in the P&C insurance industry for over 25 years with over 20 of those at Nationwide. She has experience in pricing, underwriting strategy, product management, and product development, and currently leads the team responsible for new and innovative products for all personal lines programs for Nationwide’s, industry leading SmartRide and SmartMiles UBI programs. Robin Harbage is a telematics consultant and former insurance industry executive. Robin has over 40 years of experience in roles such as chief actuary, insurance, product manager, and general manager overseeing pricing, marketing, and product development for top tier insurance carriers, most notably Progressive. And in addition, spent over a decade in consulting where he built and brought the drivability program to market and helped dozens of insurance companies launch their initial telematics offerings. He has, in fact, worked with telematics products since 1996. So let’s get to know you all a little bit better. I would love to hear from each of you some interesting innovation or application of telematics you’ve all experienced during your time working with these different products. So Gina, maybe you could kick us off.

Gina Minick:

Sure. Thanks, Louisa. I’ll share a more personal story. Louisa mentioned I’ve shared, I’ve spent my entire career in the insurance industry – about half of that in telematics. And at this point my entire family knows that I believe very strongly in safe driving and they welcome, I think, they welcome that I impose my beliefs on them by connecting all of the things in our lives, including those of my extended family. So my most enjoyable use of telematics that I’ve seen was once when my mother-in-law actually borrowed my car to drive my kids somewhere and she got home and she said, “Gina, I hate to tell you this, but

you’re going to see a hard braking event on that trip. I’m really sorry, somebody pulled out in front of me.” So I know my story’s anecdotal, but I really enjoy seeing the ability to influence people to be safer on the roads, especially when it comes to keeping my kids safe when family’s driving them.

Louisa Harbage-Edell:

Excellent. Larry, how about you?

Larry Thursby:

I don’t know if you’d call it innovative, but I always found it helpful in conversations about distracted driving to be able to refer to some of the telematics information because that’s one of those conversations that it seemed like you had every day of every year, and no one brought facts, they just brought perceptions there for a while and it was very difficult to try to shut those conversations down until you could inject actual facts. So that was helpful.

Louisa Harbage-Edell:

Absolutely. Teresa?

Teresa Scharn:

I dunno that mine’s innovative. It’s very similar to Gina’s in that the exciting thing for me on telematics is that it can make insurance mean something to customers more than what it does today. Today, people have it – have insurance because they either have to have it, and it’s almost like somewhat of a nuisance to people, to a degree. But with telematics, we have such an opportunity to create that everyday engaging experience. Fun to see my husband who is one of those people who just would never engage with things at all, but when he had that app on, he was watching his driving very, very clearly. And he is not the riskiest driver in the world, but he does make me nervous and my knuckles are white frequently. When I was driving with him, he’s the one who has to pass the car really fast. All of that stopped, and he was watching his driving, he was watching the app, he was engaging with it every day, and we see that. We see that with customers and that they engage. It’s just a tremendous opportunity for us to engage with our customers and have insurance mean something to them, have it be a benefit for them and a benefit for us. And truly, it’s a way for insurance companies to find this type of engagement or to be leading it in telematics, knowing that it’s probably going to extend to other areas in insurance as well.

Louisa Harbage-Edell:

Yeah, absolutely. Robin, how about you?

Robin Harbage:

Louisa mentioned that I’ve worked with telematics for a number of years. In fact, a very early experience was when we took in Western New York and installed 50 cars with hard-installed telematics and with mobile phones. Yes, there was a time when you actually bolted mobile phones into your car. So for those who don’t remember that time it happened, those 50 drivers then had their driving monitored and they had a call-in station where they could call in if they had questions or issues and information was sent back to them. One of our drivers driving around accidentally went off the road, flipped his car, and as he was trying to unbuckle his seatbelt, the call station actually called him up on his hard installed phone and said, “Sir, it looks like your car is parked in unusual position cause of gyroscopic capability.” They knew he was parked upside down. “Would you like us to send out some kind of help to assist you with your situation there?” So that was in the late ’90s. It was innovative at the time. It’s still, in some respects, innovative today because I think first notice of loss and safety features is one of the critical innovations of telematics. And although it’s core to rate making and the primary reason insurers wants to do it, I think for the consumer’s sake and for the industry’s sake, those things that are innovating around safety are key to telematics.

Louisa Harbage-Edell:

Yeah, excellent. So with that behind us, a little bit of background for each of you before we dive into the meat of things, I want to make sure we’re all speaking from a common understanding. So let’s lay a little groundwork and talk a little bit about some of the history of telematics and how it’s evolved for us. So Robin, we’ve obviously established that you’ve been doing this for a while. Can you talk a little bit about some of what we’ve seen to date and how telematics has evolved?

Robin Harbage:

Sure. I think one of the keys here is the technology because when we first started doing telematics, as I said, we use hard installed devices both in Buffalo, New York and in Houston, Texas at Progressive in too early tests. Those were too expensive to be programs that could go out full stream because of the cost. Then we went to install devices in the OBD. Today, the industry standard is clearly to be using mobile apps or to begin using data that’s already inherent in the vehicle because it comes from an OEM through hard solve devices at the factory. And to me, that train of telematics technology is the train of telematics because as you look at telematics, what’s made it hard to go ahead is because anything you ask a consumer to do outside of insurancey things like just filling out an app and paying their premium is really difficult to get them to do.

Louisa Harbage-Edell:

Absolutely. I mean Larry, you piloted one of the early programs as well. Is there anything you’d add to that?

Larry Thursby:

I agree a hundred percent with Robin. The only thing I think I would add is that since those early days, everybody could always say, okay, “I don’t want to go in now because it’s hardwired” – and that’s difficult for the consumer – or “I don’t want to go into the OBD-II port because there’s a lot of wastage of the devices that get sent out or not plugged in” or “I don’t want to go to the phones yet because they drain batteries.” There’s always this reason not to, and I think as an industry we’ve all had to get used to, there’s probably not a perfect end result. You might say, “Well, data’s going to come out of the car someday” and actually it already is [with connected cars], so it already is happening and that’s the end-all be-all. But I just talked to someone the other day that was working on a webcam or basically a camera feed that would compliment the computer data coming out of car so that you could eliminate those. Gina’s example of “I had a hard brake, but I really needed to have the hard brake.” So there’s always going to be someone out there innovating and coming up with a new solution that says, “Here’s the best way or the best source of data.” But I think the key is, as an industry, I see more and more people embracing the here and now of “I’ve got to get started or I’m already in this. I’ve just got to continue to evolve as things change.”

Louisa Harbage-Edell:

Gina, Teresa, anything you’d add to that?

Teresa Scharn:

The only thing I’d say is that one of the things that’s evolved over time is that we used to get data from one or two sources and now we’re getting data from multiple sources. And so the importance of being able to understand and manage data not only at scale but with that level of complexity is something that has grown and is in today’s programs as well.

Gina Minick:

And just last thought on that to Teresa’s point, that really involves all of us working together, which is why we’re here on a panel today. Because if we think about how do we use all of the history that Robin shared and think about the different ways we can collect the data and the data that’s out there in the world right now for us to harness…how do we all work together to harness that data so we can really make this happen with telematics in the industry and have more than 10% of policies signed up with telematics?

Louisa Harbage-Edell:

That’s maybe a really good segue. We’ve seen some monumental shifts this year in driving behavior in response to the COVID pandemic and companies who do have some of that critical data or who’ve been able to access that data have been able to use it to adapt. And Gina, I know you’ve been thinking a lot about that. Do you want to talk a about that a little bit?

Gina Minick:

Yeah, I’ll start with look how Covid has impacted us all. We’re on virtual meetings together. There are four virtual meetings going on in my home right now where I’m at, and consumers are driving less. So we have been tracking actual mileage to understand how are driving trends changing with COVID. That’s really been imposed on all of us across the world and we have seen notable decreases in mileage At times it’s been up to like 50% less mileage now we’re much…in a much more stable place, and miles driven year over year are getting back to a normal place. We’re still seeing people are taking fewer trips. I think, just last week, our data was showing 8% fewer trips were being taken. And we’re watching this really closely to help insurance companies make decisions on how to make rate adjustments in the future and how to keep a pulse on these trends.

But as frustrating as COVID has been for many of us, I like to try to focus on the positive changes I’ve seen. And one very promising trend is that shift in consumer mindset. I’m aware that I’m driving less and I’m aware more and in tune with how I’m driving and it’s a little more top of mind for me to think about telematics and my insurance than it was a year ago today. So Arity actually conducted research back in 2019 asking about consumer’s willingness to share driving data and different pieces of driving data, and we saw that about 30% of consumers were willing to share driving data. But then we ran that same study in May of 2020 and asked the same questions and it was fascinating to see actually there was quite a shift, more like depending on the data element, 40 to 50% willingness to share data. So this is really promising to unprecedented shifts in consumer mindset, which I like to think will open the door for insurance companies to get higher adoption of telematics.

Louisa Harbage-Edell:

You’ve of lived this experience as one of those insurance companies. Do you want to talk at all about some of what you’ve seen this year?

Teresa Scharn:

Yeah, one of the things that the telematics data has enabled really was what Gina was talking to at the beginning, and that is with Covid having an immediate understanding of the changes in mileage and the changes in how people drive and the insurance company’s reaction to that and the speed of that reaction is unprecedented. It would’ve taken us years or year at least, I would say, for those trends to be seen in our data. And if the trends came back, as Gina said, we likely would not have recognized it because trends are back to normal and you’re writing rates for the future, you’re creating rates now for the future. Well, with telematics data, we were able to see immediately how much mileage was dropping, understand exactly what that was doing to our frequencies and calculate that impact on our premiums and give those refunds back to our customers.

And many large insurance companies, many insurance companies did that within months. That is just amazing. And that was 100% enabled by telematics. Without that, nobody would’ve been comfortable doing that and we would not have had the data to do it correctly and profitably as well. So that is very, very exciting. And on the consumer interest side, we are absolutely seeing an increase in interest. We rolled out our SmartMiles, which is a pay-per-mile program last year, and we’re scaling it this year. Sometimes timing is great. And so we have seen a significant increase in the sale of that of all programs of particularly that program, it’s already 10% of our new business, which is significantly higher than we had ever anticipated, a true per-mile program being…and it’s close to 15 million a premium already this year. So exciting to see that.

Louisa Harbage-Edell:

Interesting. Excellent. Alright, so we’ve grown a lot over the past decades. This data is incredibly valuable, but we’ve also acknowledged that we do still see limitations to these programs. That there are still challenges to getting people into them to get the meaningful data necessary to get these kinds of insights. So let’s ask our audience. We have a poll question for you here around what do you see as the biggest challenge currently facing insurance companies to bring these telematics programs to the next level? Can we bring that one up? All right, so what challenges are you seeing as being significant? Is it things around distribution, customer participation, analytics, sophistication, regulation, just generally not having programs in place? We’ll just give this a minute for you all to get your answers in. It looks like those are leveling off. So I think we can go ahead and close out the poll and share the results with you.

So far and away, more than half of you feel like customer participation is the number one issue here. A few things start to filter out sort of analytics sophistication, looks like it’s coming in seconds, distribution and regulation close behind and then just not having a program in place at all certainly is still in there as an issue as well. So customer participation, I mean that’s a big one and we’ve talked a lot about this over time. I mean, I don’t know if any of you have some strong feelings about where you want to start or tackling that customer participation issue in particular.

Robin Harbage:

I can speak to that a bit. It goes to the comment I made earlier about asking customers to do something that’s not insurancey and I think it can’t be overstated that you want to make this as simple as possible for consumers. And I think one thing the audience here has to take away from this is that we’re going there, and it’s got to be a question of how fast you want to move towards this as a insurer because there is so much ambient data that is now available either through the car or through the phone, and there are people that are pulling that data together and going to be making it ready for you to use to work with your consumers. So to the extent that you can do that, so the consumer can participate simply by

saying, yes, use my data and the data’s already embedded in the insurer’s capability to pull and look at it, will really change the way that this rolls out.

Larry Thursby:

I think I would add to that, I was one of the folks early on that made the mistake of thinking, consumer interest would be high for this, but for getting that insurance in general, consumer interest is almost next to nothing. So you’ve got something that’s very interesting within a category that’s of no interest at all to 99.9% of the population. So now that I think we understand that, going back to Robin’s thought, this thing is, and something that Gina shared earlier is consumers are not averse to the concept of telematics in general. So if we can embed it within the process where they don’t have to take additional steps that are required today, I think that will make all the difference in the world and we’ll probably take that poll result to 57% consumer interest and show up for what it really is, right. Consumers, I think, will be on board as soon as it’s almost automatic to be part of the process. And I don’t think you’ll see any type of rebellion against the use of telematics from consumers, from consumer advocates or really from regulators. Certainly there are going to be questions out there, but nothing compared to what the insurance industry has gone through with other variables such as use of credit or financial responsibility and rating.

Teresa Scharn:

I think this is a case where when you ask consumers if they’re interested or you put a value prop in front of them, sometimes you get two different responses as well. And so consumers absolutely are and should be concerned about where they share their data and how they’re sharing their data. And as insurance companies, customers don’t trust us. We’re not an area where people trust. And so you have to figure out a way to get past that either by making sure you’re showing them what you’re sharing or what data they’re sharing in a very clear and concise way or by really truly having a value prop in the end when all the consumer studies we’ve done, there’s always that person who says, “I’m never going to share my data.” This isn’t going to happen. And then as soon as you say you can save up to 40% on insurance, and by the way the average is 20, I’ve seen the cybersecurity person pull up his chair and go, “What? Okay, now I’m interested.” And so it’s really in how you present it. and we changed on our front end for the direct channel that goes directly to consumers. We just changed this year how we sold it. We put little videos in explaining it to customers, did it in many different ways. And our consumer adoption went from about 60% to 80, 85% with telematics. So it really is making it simple and meaningful to them in a way that engages them. Yeah.

Gina Minick:

Larry, something you said really resonated with me around, you overestimated consumer’s interest. I think all of us have been in the industry for our careers. So when Teresa, I think we’ve all designed telematics programs and we think, “Yeah, everyone’s going to love this. I love tracking my family, I love looking at this data” – but we’re talking to ourselves. How do we take ourselves out of it and our passion for data and put ourselves in the consumer’s shoes, in the agent’s shoes or just that online flow, the call center rep, how do we make it so easy for them to get that telematics data and onboard a customer who let’s face it, may not love telematics as much as we do, right?

Louisa Harbage-Edell:

Yeah, absolutely.

Larry Thursby:

Teresa’s point, this is a story that’s fun to tell with videos or whatever that else it is and it’s a good thing to talk about. So we don’t necessarily have to shy away with it from it as we have some other things. We’re just not always that great as an industry talking to consumers about stuff like this. We’re great at handling claims, that’s what we’ve done for whatever, how many hundred years we’ve been in auto insurance. But now this is really a different thing for us to talk about. So it seems natural that it would take us a little time

Louisa Harbage-Edell:

And definitely a piece I think that requires some education. So how we do that is important. So anything else you’d add to that consumer participation or that engagement piece before we move on to one of these other issues,

Larry Thursby:

I’m going to ask Robin a question or Teresa or Gina. Robin, you mentioned making it as easy as possible for the consumer to participate in telematics. And so in my mind I go to point of sale data that’s coming from an OEM, right? So the computer’s already in the car, I’m here selling new business. I don’t need to ask you for anything other than your agreement to pull that data in and provide you to Teresa’s point, perhaps a 20%, perhaps a 40%, maybe even larger discount. Where are we at in terms of how long do you think before that’s realistic for insurance companies to use widely?

Robin Harbage:

So the issue of course with vehicles is the fact that even though models have it out there, the average age of a vehicle being 10 years older more means there’s a lot of unequipped vehicles out there and the rate of change in those vehicles slows down innovation. I’ve long said that having a car not just be a vehicle for taking in information, but the vehicle for passing information through as a modem essentially for your phone is the way to get information faster out there because then the innovation comes from the phone which goes out rather quickly. So as I think about this, I think yes, the cars will be there, they will probably be the future of this because it’ll be so rich data and it’ll be available. But in the meantime, over the next two years, phones are going to be clearly the vehicle by which we take data in and the fact that there are so many location derived apps out there on your phone already, all of which are taking a location of data and that data can be pulled together.

That’s where we’re going to see the immediate impact on telematics being adopted of eight people. I guess I’d say, and I strongly emphasize when you look at telematics or why it hasn’t taken place, it has to taken place in many respects because companies haven’t leaned into it in the hard way that a few large insurers have. And by that I mean their senior management say, yeah, we have to do it, but they’re not fully committed to it. If there’s one thing I think has prevented it, the lack of senior management buy-in and commitment to it. Because once they do and they take you to the point where it’s not just an option, it is the option, it is your default for selling insurance, that’s when you really start to push it out there. And you’ll buy a lot of, you’ll find a lot of very anxious customers who are willing to take those discounts and better rates when you put it out there if you make it the default.

Larry Thursby:

I think it’s a great point. One of the things, Robin, I think we talked about a little bit before too, the buy-in. It feels like the buy-in from executives at insurance companies for telematics is very high right now. At least what it was in the past. Everybody sees the future and what needs to happen. What I think it

feels like we have now is buy-in to do it, but not a reason. Why are you doing it now? I used this analogy the other day, it’s almost like the mask wearing, right? If you watch sports, and I’m a big sports fan, you can look at the sidelines and see the coach that wears the mask down over their chin. I call it the chin diaper and then the coach that wears it over the mouth. So there’s agreement, okay, I’ve got to wear the mask, but why am I doing it? What’s the point and am I getting at it? And I think that’s the buy-in piece that we still lack a bit, and that’s up to a lot of us to continue to stay in ways to leadership throughout the organization. Here’s why we need to be doing it now and what we get out of it now versus “Yeah, I know what we get out of it in seven years, but why am I investing, whatever it is, $500,000, a million or two million this year in telematics?”

Louisa Harbage-Edell:

And interestingly from our audience that internal buy-in was one of the lowest points of concern. So I think we really have seen and increase in that buy-in and a shift in that area. So the second one that really came up for people was that distribution issue. So Teresa, I know you’ve put a lot of thought into this, you want to talk a little bit about maybe how you’ve overcome some of those distribution challenges.

Teresa Scharn:

And I think I’d say we’re all still overcoming some of those distribution challenges. So we’ve made a lot of headway. So we are largely independent agents or solely independent agents for the most part. We do have a direct channel now and it is very challenging and some things that honestly we could do better. And the same message goes for independent agents as what we just said for customers. You have to make it simple. This is new, this is different, this is changing the way they sell insurance. And that is when it’s confusing and hard, that’s when it just doesn’t get done. And so even having multiple programs, having multiple versions of programs, having multiple…having our programs side by side with a Progressive program right inside of an agency – that’s confusing for agents. And I’d say the nut that we’re trying to crack and still trying to crack is really how do we keep it as simple as possible at point of sale for agents and make it really easy for them to work through the process?

The piece that I think we have done well is helping them understand some of the why and what’s in it for them, which is why they’re selling it. So the 10% participation discount alone from a savings perspective is huge. And then we also have an option where they can get the full participation discount or the full telematics discount if they have data that comes with them so that the competitive pricing upfront, alone helps them tremendously from a sales perspective and then the retention side, understanding that they spend a lot of time to get business in the door, they want to keep it, they don’t want to lose that customer. And so the impacts that we have on retention and making sure helping them understand that I think we’ve gotten past that. That message has gotten out fairly well, but we have room to grow in making it simple.

Louisa Harbage-Edell:

Absolutely. I mean, Robin, Gina, you’ve helped roll out a number of programs and I know that’s been a key concern. Anything you’d add in your thinking there?

Gina Minick:

I think of you threw up many challenges in that question, Louisa said, I think of it as like a progression or peeling back the onion. At the end of the day, the insurance industry wants this data because we want to understand how to price people accurately. Insurance is about understanding risks. So we want the

data to understand risk, but we have to get the data by creating an app, creating an engaging experience and getting that consumer data. But to Teresa’s point, we can’t get that data unless we have effective distribution and we make it so easy. So as we keep peeling it back, we think, well, we know what we want problem trying to solve, but it has to start with a simple program in a simple way for agents or again, your online flow to just seamlessly have somebody push through and got to keep in mind insurance companies are trying to bind policies too. So it has to be simple so that you get from quotes to bind and then through to actually get that telematics data so we can get what we need at the end of the day, which is access to that data to better understand the risk.

Louisa Harbage-Edell:

So I think the last of the top three was really around this idea of analytics sophistication and how we think about that, especially when you’ve got this kind of data and all of these variables. Robin, I mean you’ve obviously put a lot of thought into some of these programs and getting to that point. Anything you’d throw on the table with regards to that one?

Robin Harbage:

Certainly. So the sophistication has largely been hamstrung by the fact that all insurers have used this as a discount tacked on to the rest of their rating variables. So if you really want to make this something of import and to use as full power, and let’s be honest, it has the most power of any rating variable that we’ve discovered so far, there’s no question about that. So if you want to make use of that and in fact reduce the reliance on some of the proxies that have some problems, if you will, societally as well as from the standpoint of the insurer’s acceptance of some of those variables like credit, you’d have to say, look, we’re going to make this a part of our generalized linear model. We’re going to just put it in as its full power and let it take off all the handcuffs that we’ve been on it because we used it only on the residual of signal that’s not identified for other proxy variables.

Once you do that and you make it core to your rating, its power will be tremendous. It’ll attract all the best drivers. It’ll segment that company’s data better than anybody else. It’ll be in a kind of innovation that has largely driven some major insurers’ growth for years and years. I’d be surprised if some insurers aren’t within a year away from making that a core to their product and will be gone – the other insurers that don’t have that power once it starts happening. In fact, there are some insurers already out there who have made telematics the core of the entire rating. So I think that largely is where insurers have to go and have to go quickly.

Larry Thursby:

I think the statement you just made, Robin is scary, right? And Nationwide, Teresa has carried this on and done a great job. I think one of the purposes we implemented telematics early on was to get data so that we could do the analysis so that we could be ready when the class plan revolution occurred to make the pivot the same time as some of the true leaders that are out there. And the fact that in the poll there was a large percentage of folks who said analytics sophistication is a concern in terms of its telematics, continued evolution tells you that yes, those carriers who have the data and are ready to begin using telematics as something other than a monovariate discount, take it to the top of your rating equation and vary other things like age and gender and driving points – all those other things begin taking away from those and adding to telematics.

That should be scary to everyone. And I think the question that it raised, at least for me from a Nationwide perspective is, can we as a company go fast enough on our own with our own data, doing our own warehousing, doing our own analytics to build that? Or are we forced to go into with a third party like Arity, like Lexis, some of the other folks who are aggregating even more data so that we are ready to go sooner rather than later? And Teresa, I don’t know if you have any perspective, if you continue to stay the course on we’re going to kind of get all our data, we’re going to do it ourselves or are we going to form that piece out?

Teresa Scharn:

Yeah, and I think we’re always thinking about what’s the most efficient and most effective way to bring this to life. And as we think through that, one of the number one things that I think anybody needs to think about is where do you want to differentiate? Where do you want to differentiate your program and be very clear and be very focused on that. And then why do you want to differentiate on that? And I’ll give you an example to that, going back to this point is that when you think about analytics, there’s lots of steps to this. There’s the data integration, there’s just the data management of all of that data in which you get it in in-house, the data science behind taking that data, normalizing it, making all the data features, and then there’s the modeling on top of it. And this is at a scale that we haven’t seen before.

So which part of each one of those steps is important to be really fast at, which part of these steps is really important to have be Nationwide only, which part isn’t? And I think you can partner with any piece and part of that. Today, Nationwide does do all of that ourselves, and we do a lot of it in-house now. We didn’t start in-house, we started by partnering with a company with Robin actually at the very beginning, right Larry, where we started partnering and used their scoring and their rating, but ended up building it all. But I would say that’s all TBD to bring in data very fast to integrate with new customers faster and new data sources faster. Will we be able to do that better if we partner with somebody versus build it ourselves? That’s something we’re actually asking right now and actually have an RFP out right now to figure out that solution. And it’s because we understand what our goal is, we understand where we want to differentiate and what the end solution needs to be and it might be better to partner to do that.

Larry Thursby:

And just so folks don’t feel like they’re doing something wrong, that’s hard. But I can remember when we started, we had a minimal number of customers in the program and we couldn’t get the data across communication lines into our warehouse. That took us months, then we got it into a warehouse and we couldn’t figure out how to get the data out of the warehouse to actually begin to analyze it and study it. And that was with one source. And now as you’ve said previously, you’re getting data from multiple cell phone, third party providers from OBD-II, from OEMs, and the amount of data is so much larger. That is a task that’s monumental for just about any insurance company, no matter how good they are at data science. So if there’s folks out there struggling with it, they shouldn’t feel like they’re alone. Everybody I think save maybe one or two struggles with that.

Gina Minick:

And I think this does apply to all insurance companies regardless of size. It’s like how do you find the right recipe for you based on where your journey is on telematics, what are the things that you need help from other vendors on because you don’t have the capabilities because you don’t have them yet because you’ll never build ’em because it’s not core to your tech stack and it’s something you don’t need to invest in. How do you balance that level of investment through the whole data flow, all the way f

Louisa Harbage-Edell:

“Telematics will revolutionize the insurance industry.” How many of you have heard this bold statement, and how many of you are still waiting for that meaningful change to really come to fruition five, 10, or even 15 years later? My name is Louisa Harbage-Edell, and today I’m joined by four industry experts for this bootcamp on innovating the future of auto insurance. We’ve reached a point where a large number of companies have a telematics program in place, but many still struggle to gain critical mass. In fact, less than 10% of policies enforced today across the industry are usage-based auto insurance ones. So what’s holding us back, and how can we get past these barriers to achieve the opportunities we see for telematics in the future? In today’s informal discussion, my panelists will address these issues and more. We’ll also be asking you, our audience, some poll questions to get your thoughts, and we’ll open it up for your questions via live Q&A towards the end of the session. That said, you’ll see that Q&A panel is open throughout, so feel free to drop in your questions at any time. By the end of this conversation, you’ll have ideas for how to overcome some of the key challenges still facing companies today in order to take your program to the next level, as well as key insight into the exciting new ways telematics is evolving for the future. With that, let me introduce our panelists.

First up, Gina Minick is product director for Insurance Solutions at Arity. She has extensive experience in product management as well as, prior to Arity, being at Allstate, where she played an integral role in the countrywide rollout of Drivewise, Allstate’s telematics program. She’s also served in numerous other roles at Allstate and Progressive, including product strategy and implementation, risk management, catastrophe management, competitive intelligence and marketing. Larry Thursby is a P&C insurance consultant and a former Nationwide executive. He worked at Nationwide for 25 years before retiring in 2017. His final 10 years at the company were as vice president of personalized product and pricing, and it was of course during this time that Nationwide introduced their first UBI program for consumers. After initial success in that pilot state, that offering grew into the countrywide program including OBD-II [On-Board Diagnostic II] and smartphone connections as well as connected car solutions.

Teresa Scharn is the current VP for personal lines product development at Nationwide and has worked in the P&C insurance industry for over 25 years with over 20 of those at Nationwide. She has experience in pricing, underwriting strategy, product management, and product development, and currently leads the team responsible for new and innovative products for all personal lines programs for Nationwide’s, industry leading SmartRide and SmartMiles UBI programs. Robin Harbage is a telematics consultant and former insurance industry executive. Robin has over 40 years of experience in roles such as chief actuary, insurance, product manager, and general manager overseeing pricing, marketing, and product development for top tier insurance carriers, most notably Progressive. And in addition, spent over a decade in consulting where he built and brought the drivability program to market and helped dozens of insurance companies launch their initial telematics offerings. He has, in fact, worked with telematics products since 1996. So let’s get to know you all a little bit better. I would love to hear from each of you some interesting innovation or application of telematics you’ve all experienced during your time working with these different products. So Gina, maybe you could kick us off.

Gina Minick:

Sure. Thanks, Louisa. I’ll share a more personal story. Louisa mentioned I’ve shared, I’ve spent my entire career in the insurance industry – about half of that in telematics. And at this point my entire family knows that I believe very strongly in safe driving and they welcome, I think, they welcome that I impose my beliefs on them by connecting all of the things in our lives, including those of my extended family. So my most enjoyable use of telematics that I’ve seen was once when my mother-in-law actually borrowed my car to drive my kids somewhere and she got home and she said, “Gina, I hate to tell you this, but you’re going to see a hard braking event on that trip. I’m really sorry, somebody pulled out in front of me.” So I know my story’s anecdotal, but I really enjoy seeing the ability to influence people to be safer on the roads, especially when it comes to keeping my kids safe when family’s driving them.

Louisa Harbage-Edell:

Excellent. Larry, how about you?

Larry Thursby:

I don’t know if you’d call it innovative, but I always found it helpful in conversations about distracted driving to be able to refer to some of the telematics information because that’s one of those conversations that it seemed like you had every day of every year, and no one brought facts, they just brought perceptions there for a while and it was very difficult to try to shut those conversations down until you could inject actual facts. So that was helpful.

Louisa Harbage-Edell:

Absolutely. Teresa?

Teresa Scharn:

I dunno that mine’s innovative. It’s very similar to Gina’s in that the exciting thing for me on telematics is that it can make insurance mean something to customers more than what it does today. Today, people have it – have insurance because they either have to have it, and it’s almost like somewhat of a nuisance to people, to a degree. But with telematics, we have such an opportunity to create that everyday engaging experience. Fun to see my husband who is one of those people who just would never engage with things at all, but when he had that app on, he was watching his driving very, very clearly. And he is not the riskiest driver in the world, but he does make me nervous and my knuckles are white frequently. When I was driving with him, he’s the one who has to pass the car really fast. All of that stopped, and he was watching his driving, he was watching the app, he was engaging with it every day, and we see that. We see that with customers and that they engage. It’s just a tremendous opportunity for us to engage with our customers and have insurance mean something to them, have it be a benefit for them and a benefit for us. And truly, it’s a way for insurance companies to find this type of engagement or to be leading it in telematics, knowing that it’s probably going to extend to other areas in insurance as well.

Louisa Harbage-Edell:

Yeah, absolutely. Robin, how about you?

Robin Harbage:

Louisa mentioned that I’ve worked with telematics for a number of years. In fact, a very early experience was when we took in Western New York and installed 50 cars with hard-installed telematics and with mobile phones. Yes, there was a time when you actually bolted mobile phones into your car. So for those who don’t remember that time it happened, those 50 drivers then had their driving monitored and they had a call-in station where they could call in if they had questions or issues and information was sent back to them. One of our drivers driving around accidentally went off the road, flipped his car, and as he was trying to unbuckle his seatbelt, the call station actually called him up on his hard installed phone and said, “Sir, it looks like your car is parked in unusual position cause of gyroscopic capability.” They knew he was parked upside down. “Would you like us to send out some kind of help to assist you with your situation there?” So that was in the late ’90s. It was innovative at the time. It’s still, in some respects, innovative today because I think first notice of loss and safety features is one of the critical innovations of telematics. And although it’s core to rate making and the primary reason insurers wants to do it, I think for the consumer’s sake and for the industry’s sake, those things that are innovating around safety are key to telematics.

Louisa Harbage-Edell:

Yeah, excellent. So with that behind us, a little bit of background for each of you before we dive into the meat of things, I want to make sure we’re all speaking from a common understanding. So let’s lay a little groundwork and talk a little bit about some of the history of telematics and how it’s evolved for us. So Robin, we’ve obviously established that you’ve been doing this for a while. Can you talk a little bit about some of what we’ve seen to date and how telematics has evolved?

Robin Harbage:

Sure. I think one of the keys here is the technology because when we first started doing telematics, as I said, we use hard installed devices both in Buffalo, New York and in Houston, Texas at Progressive in too early tests. Those were too expensive to be programs that could go out full stream because of the cost. Then we went to install devices in the OBD. Today, the industry standard is clearly to be using mobile apps or to begin using data that’s already inherent in the vehicle because it comes from an OEM through hard solve devices at the factory. And to me, that train of telematics technology is the train of telematics because as you look at telematics, what’s made it hard to go ahead is because anything you ask a consumer to do outside of insurancey things like just filling out an app and paying their premium is really difficult to get them to do.

Louisa Harbage-Edell:

Absolutely. I mean Larry, you piloted one of the early programs as well. Is there anything you’d add to that?

Larry Thursby:

I agree a hundred percent with Robin. The only thing I think I would add is that since those early days, everybody could always say, okay, “I don’t want to go in now because it’s hardwired” – and that’s difficult for the consumer – or “I don’t want to go into the OBD-II port because there’s a lot of wastage of the devices that get sent out or not plugged in” or “I don’t want to go to the phones yet because they drain batteries.” There’s always this reason not to, and I think as an industry we’ve all had to get used to, there’s probably not a perfect end result. You might say, “Well, data’s going to come out of the car someday” and actually it already is [with connected cars], so it already is happening and that’s the end-all be-all. But I just talked to someone the other day that was working on a webcam or basically a camera feed that would compliment the computer data coming out of car so that you could eliminate those. Gina’s example of “I had a hard brake, but I really needed to have the hard brake.” So there’s always going to be someone out there innovating and coming up with a new solution that says, “Here’s the best way or the best source of data.” But I think the key is, as an industry, I see more and more people embracing the here and now of “I’ve got to get started or I’m already in this. I’ve just got to continue to evolve as things change.”

Louisa Harbage-Edell:

Gina, Teresa, anything you’d add to that?

Teresa Scharn:

The only thing I’d say is that one of the things that’s evolved over time is that we used to get data from one or two sources and now we’re getting data from multiple sources. And so the importance of being able to understand and manage data not only at scale but with that level of complexity is something that has grown and is in today’s programs as well.

Gina Minick:

And just last thought on that to Teresa’s point, that really involves all of us working together, which is why we’re here on a panel today. Because if we think about how do we use all of the history that Robin shared and think about the different ways we can collect the data and the data that’s out there in the world right now for us to harness…how do we all work together to harness that data so we can really make this happen with telematics in the industry and have more than 10% of policies signed up with telematics?

Louisa Harbage-Edell:

That’s maybe a really good segue. We’ve seen some monumental shifts this year in driving behavior in response to the COVID pandemic and companies who do have some of that critical data or who’ve been able to access that data have been able to use it to adapt. And Gina, I know you’ve been thinking a lot about that. Do you want to talk a about that a little bit?

Gina Minick:

Yeah, I’ll start with look how Covid has impacted us all. We’re on virtual meetings together. There are four virtual meetings going on in my home right now where I’m at, and consumers are driving less. So we have been tracking actual mileage to understand how are driving trends changing with COVID. That’s really been imposed on all of us across the world and we have seen notable decreases in mileage At times it’s been up to like 50% less mileage now we’re much…in a much more stable place, and miles driven year over year are getting back to a normal place. We’re still seeing people are taking fewer trips. I think, just last week, our data was showing 8% fewer trips were being taken. And we’re watching this really closely to help insurance companies make decisions on how to make rate adjustments in the future and how to keep a pulse on these trends.

But as frustrating as COVID has been for many of us, I like to try to focus on the positive changes I’ve seen. And one very promising trend is that shift in consumer mindset. I’m aware that I’m driving less and I’m aware more and in tune with how I’m driving and it’s a little more top of mind for me to think about telematics and my insurance than it was a year ago today. So Arity actually conducted research back in 2019 asking about consumer’s willingness to share driving data and different pieces of driving data, and we saw that about 30% of consumers were willing to share driving data. But then we ran that same study in May of 2020 and asked the same questions and it was fascinating to see actually there was quite a shift, more like depending on the data element, 40 to 50% willingness to share data. So this is really promising to unprecedented shifts in consumer mindset, which I like to think will open the door for insurance companies to get higher adoption of telematics.

Louisa Harbage-Edell:

You’ve of lived this experience as one of those insurance companies. Do you want to talk at all about some of what you’ve seen this year?

Teresa Scharn:

Yeah, one of the things that the telematics data has enabled really was what Gina was talking to at the beginning, and that is with Covid having an immediate understanding of the changes in mileage and the changes in how people drive and the insurance company’s reaction to that and the speed of that reaction is unprecedented. It would’ve taken us years or year at least, I would say, for those trends to be seen in our data. And if the trends came back, as Gina said, we likely would not have recognized it because trends are back to normal and you’re writing rates for the future, you’re creating rates now for the future. Well, with telematics data, we were able to see immediately how much mileage was dropping, understand exactly what that was doing to our frequencies and calculate that impact on our premiums and give those refunds back to our customers.

And many large insurance companies, many insurance companies did that within months. That is just amazing. And that was 100% enabled by telematics. Without that, nobody would’ve been comfortable doing that and we would not have had the data to do it correctly and profitably as well. So that is very, very exciting. And on the consumer interest side, we are absolutely seeing an increase in interest. We rolled out our SmartMiles, which is a pay-per-mile program last year, and we’re scaling it this year. Sometimes timing is great. And so we have seen a significant increase in the sale of that of all programs of particularly that program, it’s already 10% of our new business, which is significantly higher than we had ever anticipated, a true per-mile program being…and it’s close to 15 million a premium already this year. So exciting to see that.

Louisa Harbage-Edell:

Interesting. Excellent. Alright, so we’ve grown a lot over the past decades. This data is incredibly valuable, but we’ve also acknowledged that we do still see limitations to these programs. That there are still challenges to getting people into them to get the meaningful data necessary to get these kinds of insights. So let’s ask our audience. We have a poll question for you here around what do you see as the biggest challenge currently facing insurance companies to bring these telematics programs to the next level? Can we bring that one up? All right, so what challenges are you seeing as being significant? Is it things around distribution, customer participation, analytics, sophistication, regulation, just generally not having programs in place? We’ll just give this a minute for you all to get your answers in. It looks like those are leveling off. So I think we can go ahead and close out the poll and share the results with you.

So far and away, more than half of you feel like customer participation is the number one issue here. A few things start to filter out sort of analytics sophistication, looks like it’s coming in seconds, distribution and regulation close behind and then just not having a program in place at all certainly is still in there as an issue as well. So customer participation, I mean that’s a big one and we’ve talked a lot about this over time. I mean, I don’t know if any of you have some strong feelings about where you want to start or tackling that customer participation issue in particular.

Robin Harbage:

I can speak to that a bit. It goes to the comment I made earlier about asking customers to do something that’s not insurancey and I think it can’t be overstated that you want to make this as simple as possible for consumers. And I think one thing the audience here has to take away from this is that we’re going there, and it’s got to be a question of how fast you want to move towards this as a insurer because there is so much ambient data that is now available either through the car or through the phone, and there are people that are pulling that data together and going to be making it ready for you to use to work with your consumers. So to the extent that you can do that, so the consumer can participate simply by saying, yes, use my data and the data’s already embedded in the insurer’s capability to pull and look at it, will really change the way that this rolls out.

Larry Thursby:

I think I would add to that, I was one of the folks early on that made the mistake of thinking, consumer interest would be high for this, but for getting that insurance in general, consumer interest is almost next to nothing. So you’ve got something that’s very interesting within a category that’s of no interest at all to 99.9% of the population. So now that I think we understand that, going back to Robin’s thought, this thing is, and something that Gina shared earlier is consumers are not averse to the concept of telematics in general. So if we can embed it within the process where they don’t have to take additional steps that are required today, I think that will make all the difference in the world and we’ll probably take that poll result to 57% consumer interest and show up for what it really is, right. Consumers, I think, will be on board as soon as it’s almost automatic to be part of the process. And I don’t think you’ll see any type of rebellion against the use of telematics from consumers, from consumer advocates or really from regulators. Certainly there are going to be questions out there, but nothing compared to what the insurance industry has gone through with other variables such as use of credit or financial responsibility and rating.

Teresa Scharn:

I think this is a case where when you ask consumers if they’re interested or you put a value prop in front of them, sometimes you get two different responses as well. And so consumers absolutely are and should be concerned about where they share their data and how they’re sharing their data. And as insurance companies, customers don’t trust us. We’re not an area where people trust. And so you have to figure out a way to get past that either by making sure you’re showing them what you’re sharing or what data they’re sharing in a very clear and concise way or by really truly having a value prop in the end when all the consumer studies we’ve done, there’s always that person who says, “I’m never going to share my data.” This isn’t going to happen. And then as soon as you say you can save up to 40% on insurance, and by the way the average is 20, I’ve seen the cybersecurity person pull up his chair and go, “What? Okay, now I’m interested.” And so it’s really in how you present it. and we changed on our front end for the direct channel that goes directly to consumers. We just changed this year how we sold it. We put little videos in explaining it to customers, did it in many different ways. And our consumer adoption went from about 60% to 80, 85% with telematics. So it really is making it simple and meaningful to them in a way that engages them. Yeah.

Gina Minick:

Larry, something you said really resonated with me around, you overestimated consumer’s interest. I think all of us have been in the industry for our careers. So when Teresa, I think we’ve all designed telematics programs and we think, “Yeah, everyone’s going to love this. I love tracking my family, I love looking at this data” – but we’re talking to ourselves. How do we take ourselves out of it and our passion for data and put ourselves in the consumer’s shoes, in the agent’s shoes or just that online flow, the call center rep, how do we make it so easy for them to get that telematics data and onboard a customer who let’s face it, may not love telematics as much as we do, right?

Louisa Harbage-Edell:

Yeah, absolutely.

Larry Thursby:

Teresa’s point, this is a story that’s fun to tell with videos or whatever that else it is and it’s a good thing to talk about. So we don’t necessarily have to shy away with it from it as we have some other things. We’re just not always that great as an industry talking to consumers about stuff like this. We’re great at handling claims, that’s what we’ve done for whatever, how many hundred years we’ve been in auto insurance. But now this is really a different thing for us to talk about. So it seems natural that it would take us a little time

Louisa Harbage-Edell:

And definitely a piece I think that requires some education. So how we do that is important. So anything else you’d add to that consumer participation or that engagement piece before we move on to one of these other issues.

Larry Thursby:

I’m going to ask Robin a question or Teresa or Gina. Robin, you mentioned making it as easy as possible for the consumer to participate in telematics. And so in my mind I go to point of sale data that’s coming from an OEM, right? So the computer’s already in the car, I’m here selling new business. I don’t need to ask you for anything other than your agreement to pull that data in and provide you to Teresa’s point, perhaps a 20%, perhaps a 40%, maybe even larger discount. Where are we at in terms of how long do you think before that’s realistic for insurance companies to use widely?

Robin Harbage:

So the issue of course with vehicles is the fact that even though models have it out there, the average age of a vehicle being 10 years older more means there’s a lot of unequipped vehicles out there and the rate of change in those vehicles slows down innovation. I’ve long said that having a car not just be a vehicle for taking in information, but the vehicle for passing information through as a modem essentially for your phone is the way to get information faster out there because then the innovation comes from the phone which goes out rather quickly. So as I think about this, I think yes, the cars will be there, they will probably be the future of this because it’ll be so rich data and it’ll be available. But in the meantime, over the next two years, phones are going to be clearly the vehicle by which we take data in and the fact that there are so many location derived apps out there on your phone already, all of which are taking a location of data and that data can be pulled together.

That’s where we’re going to see the immediate impact on telematics being adopted of eight people. I guess I’d say, and I strongly emphasize when you look at telematics or why it hasn’t taken place, it has to taken place in many respects because companies haven’t leaned into it in the hard way that a few large insurers have. And by that I mean their senior management say, yeah, we have to do it, but they’re not fully committed to it. If there’s one thing I think has prevented it, the lack of senior management buy-in and commitment to it. Because once they do and they take you to the point where it’s not just an option, it is the option, it is your default for selling insurance, that’s when you really start to push it out there. And you’ll buy a lot of, you’ll find a lot of very anxious customers who are willing to take those discounts and better rates when you put it out there if you make it the default.

Larry Thursby:

I think it’s a great point. One of the things, Robin, I think we talked about a little bit before too, the buy-in. It feels like the buy-in from executives at insurance companies for telematics is very high right now. At least what it was in the past. Everybody sees the future and what needs to happen. What I think it feels like we have now is buy-in to do it, but not a reason. Why are you doing it now? I used this analogy the other day, it’s almost like the mask wearing, right? If you watch sports, and I’m a big sports fan, you can look at the sidelines and see the coach that wears the mask down over their chin. I call it the chin diaper and then the coach that wears it over the mouth. So there’s agreement, okay, I’ve got to wear the mask, but why am I doing it? What’s the point and am I getting at it? And I think that’s the buy-in piece that we still lack a bit, and that’s up to a lot of us to continue to stay in ways to leadership throughout the organization. Here’s why we need to be doing it now and what we get out of it now versus “Yeah, I know what we get out of it in seven years, but why am I investing, whatever it is, $500,000, a million or two million this year in telematics?”

Louisa Harbage-Edell:

And interestingly from our audience that internal buy-in was one of the lowest points of concern. So I think we really have seen and increase in that buy-in and a shift in that area. So the second one that really came up for people was that distribution issue. So Teresa, I know you’ve put a lot of thought into this, you want to talk a little bit about maybe how you’ve overcome some of those distribution challenges.

Teresa Scharn:

And I think I’d say we’re all still overcoming some of those distribution challenges. So we’ve made a lot of headway. So we are largely independent agents or solely independent agents for the most part. We do have a direct channel now and it is very challenging and some things that honestly we could do better. And the same message goes for independent agents as what we just said for customers. You have to make it simple. This is new, this is different, this is changing the way they sell insurance. And that is when it’s confusing and hard, that’s when it just doesn’t get done. And so even having multiple programs, having multiple versions of programs, having multiple…having our programs side by side with a Progressive program right inside of an agency – that’s confusing for agents. And I’d say the nut that we’re trying to crack and still trying to crack is really how do we keep it as simple as possible at point of sale for agents and make it really easy for them to work through the process?

The piece that I think we have done well is helping them understand some of the why and what’s in it for them, which is why they’re selling it. So the 10% participation discount alone from a savings perspective is huge. And then we also have an option where they can get the full participation discount or the full telematics discount if they have data that comes with them so that the competitive pricing upfront, alone helps them tremendously from a sales perspective and then the retention side, understanding that they spend a lot of time to get business in the door, they want to keep it, they don’t want to lose that customer. And so the impacts that we have on retention and making sure helping them understand that I think we’ve gotten past that. That message has gotten out fairly well, but we have room to grow in making it simple.

Louisa Harbage-Edell:

Absolutely. I mean, Robin, Gina, you’ve helped roll out a number of programs and I know that’s been a key concern. Anything you’d add in your thinking there?

Gina Minick:

I think of you threw up many challenges in that question, Louisa said, I think of it as like a progression or peeling back the onion. At the end of the day, the insurance industry wants this data because we want to understand how to price people accurately. Insurance is about understanding risks. So we want the data to understand risk, but we have to get the data by creating an app, creating an engaging experience and getting that consumer data. But to Teresa’s point, we can’t get that data unless we have effective distribution and we make it so easy. So as we keep peeling it back, we think, well, we know what we want problem trying to solve, but it has to start with a simple program in a simple way for agents or again, your online flow to just seamlessly have somebody push through and got to keep in mind insurance companies are trying to bind policies too. So it has to be simple so that you get from quotes to bind and then through to actually get that telematics data so we can get what we need at the end of the day, which is access to that data to better understand the risk.

Louisa Harbage-Edell:

So I think the last of the top three was really around this idea of analytics sophistication and how we think about that, especially when you’ve got this kind of data and all of these variables. Robin, I mean you’ve obviously put a lot of thought into some of these programs and getting to that point. Anything you’d throw on the table with regards to that one?

Robin Harbage:

Certainly. So the sophistication has largely been hamstrung by the fact that all insurers have used this as a discount tacked on to the rest of their rating variables. So if you really want to make this something of import and to use as full power, and let’s be honest, it has the most power of any rating variable that we’ve discovered so far, there’s no question about that. So if you want to make use of that and in fact reduce the reliance on some of the proxies that have some problems, if you will, societally as well as from the standpoint of the insurer’s acceptance of some of those variables like credit, you’d have to say, look, we’re going to make this a part of our generalized linear model. We’re going to just put it in as its full power and let it take off all the handcuffs that we’ve been on it because we used it only on the residual of signal that’s not identified for other proxy variables.

Once you do that and you make it core to your rating, its power will be tremendous. It’ll attract all the best drivers. It’ll segment that company’s data better than anybody else. It’ll be in a kind of innovation that has largely driven some major insurers’ growth for years and years. I’d be surprised if some insurers aren’t within a year away from making that a core to their product and will be gone – the other insurers that don’t have that power once it starts happening. In fact, there are some insurers already out there who have made telematics the core of the entire rating. So I think that largely is where insurers have to go and have to go quickly.

Larry Thursby:

I think the statement you just made, Robin is scary, right? And Nationwide, Teresa has carried this on and done a great job. I think one of the purposes we implemented telematics early on was to get data so that we could do the analysis so that we could be ready when the class plan revolution occurred to make the pivot the same time as some of the true leaders that are out there. And the fact that in the poll there was a large percentage of folks who said analytics sophistication is a concern in terms of its telematics, continued evolution tells you that yes, those carriers who have the data and are ready to begin using telematics as something other than a monovariate discount, take it to the top of your rating equation and vary other things like age and gender and driving points – all those other things begin taking away from those and adding to telematics.

That should be scary to everyone. And I think the question that it raised, at least for me from a Nationwide perspective is, can we as a company go fast enough on our own with our own data, doing our own warehousing, doing our own analytics to build that? Or are we forced to go into with a third party like Arity, like Lexis, some of the other folks who are aggregating even more data so that we are ready to go sooner rather than later? And Teresa, I don’t know if you have any perspective, if you continue to stay the course on we’re going to kind of get all our data, we’re going to do it ourselves or are we going to form that piece out?

Teresa Scharn:

Yeah, and I think we’re always thinking about what’s the most efficient and most effective way to bring this to life. And as we think through that, one of the number one things that I think anybody needs to think about is where do you want to differentiate? Where do you want to differentiate your program and be very clear and be very focused on that. And then why do you want to differentiate on that? And I’ll give you an example to that, going back to this point is that when you think about analytics, there’s lots of steps to this. There’s the data integration, there’s just the data management of all of that data in which you get it in in-house, the data science behind taking that data, normalizing it, making all the data features, and then there’s the modeling on top of it. And this is at a scale that we haven’t seen before.

So which part of each one of those steps is important to be really fast at, which part of these steps is really important to have be Nationwide only, which part isn’t? And I think you can partner with any piece and part of that. Today, Nationwide does do all of that ourselves, and we do a lot of it in-house now. We didn’t start in-house, we started by partnering with a company with Robin actually at the very beginning, right Larry, where we started partnering and used their scoring and their rating, but ended up building it all. But I would say that’s all TBD to bring in data very fast to integrate with new customers faster and new data sources faster. Will we be able to do that better if we partner with somebody versus build it ourselves? That’s something we’re actually asking right now and actually have an RFP out right now to figure out that solution. And it’s because we understand what our goal is, we understand where we want to differentiate and what the end solution needs to be and it might be better to partner to do that.

Larry Thursby:

And just so folks don’t feel like they’re doing something wrong, that’s hard. But I can remember when we started, we had a minimal number of customers in the program and we couldn’t get the data across communication lines into our warehouse. That took us months, then we got it into a warehouse and we couldn’t figure out how to get the data out of the warehouse to actually begin to analyze it and study it. And that was with one source. And now as you’ve said previously, you’re getting data from multiple cell phone, third party providers from OBD-II, from OEMs, and the amount of data is so much larger. That is a task that’s monumental for just about any insurance company, no matter how good they are at data science. So if there’s folks out there struggling with it, they shouldn’t feel like they’re alone. Everybody I think save maybe one or two struggles with that.

Gina Minick:

And I think this does apply to all insurance companies regardless of size. It’s like how do you find the right recipe for you based on where your journey is on telematics, what are the things that you need help from other vendors on because you don’t have the capabilities because you don’t have them yet because you’ll never build ’em because it’s not core to your tech stack and it’s something you don’t need to invest in. How do you balance that level of investment through the whole data flow, all the way f

Louisa Harbage-Edell:

“Telematics will revolutionize the insurance industry.” How many of you have heard this bold statement, and how many of you are still waiting for that meaningful change to really come to fruition five, 10, or even 15 years later? My name is Louisa Harbage-Edell, and today I’m joined by four industry experts for this bootcamp on innovating the future of auto insurance. We’ve reached a point where a large number of companies have a telematics program in place, but many still struggle to gain critical mass. In fact, less than 10% of policies enforced today across the industry are usage-based auto insurance ones. So what’s holding us back, and how can we get past these barriers to achieve the opportunities we see for telematics in the future? In today’s informal discussion, my panelists will address these issues and more. We’ll also be asking you, our audience, some poll questions to get your thoughts, and we’ll open it up for your questions via live Q&A towards the end of the session. That said, you’ll see that Q&A panel is open throughout, so feel free to drop in your questions at any time. By the end of this conversation, you’ll have ideas for how to overcome some of the key challenges still facing companies today in order to take your program to the next level, as well as key insight into the exciting new ways telematics is evolving for the future. With that, let me introduce our panelists.

First up, Gina Minick is product director for Insurance Solutions at Arity. She has extensive experience in product management as well as, prior to Arity, being at Allstate, where she played an integral role in the countrywide rollout of Drivewise, Allstate’s telematics program. She’s also served in numerous other roles at Allstate and Progressive, including product strategy and implementation, risk management, catastrophe management, competitive intelligence and marketing. Larry Thursby is a P&C insurance consultant and a former Nationwide executive. He worked at Nationwide for 25 years before retiring in 2017. His final 10 years at the company were as vice president of personalized product and pricing, and it was of course during this time that Nationwide introduced their first UBI program for consumers. After initial success in that pilot state, that offering grew into the countrywide program including OBD-II [On-Board Diagnostic II] and smartphone connections as well as connected car solutions.

Teresa Scharn is the current VP for personal lines product development at Nationwide and has worked in the P&C insurance industry for over 25 years with over 20 of those at Nationwide. She has experience in pricing, underwriting strategy, product management, and product development, and currently leads the team responsible for new and innovative products for all personal lines programs for Nationwide’s, industry leading SmartRide and SmartMiles UBI programs. Robin Harbage is a telematics consultant and former insurance industry executive. Robin has over 40 years of experience in roles such as chief actuary, insurance, product manager, and general manager overseeing pricing, marketing, and product development for top tier insurance carriers, most notably Progressive. And in addition, spent over a decade in consulting where he built and brought the drivability program to market and helped dozens of insurance companies launch their initial telematics offerings. He has, in fact, worked with telematics products since 1996. So let’s get to know you all a little bit better. I would love to hear from each of you some interesting innovation or application of telematics you’ve all experienced during your time working with these different products. So Gina, maybe you could kick us off.

Gina Minick:

Sure. Thanks, Louisa. I’ll share a more personal story. Louisa mentioned I’ve shared, I’ve spent my entire career in the insurance industry – about half of that in telematics. And at this point my entire family knows that I believe very strongly in safe driving and they welcome, I think, they welcome that I impose my beliefs on them by connecting all of the things in our lives, including those of my extended family. So my most enjoyable use of telematics that I’ve seen was once when my mother-in-law actually borrowed my car to drive my kids somewhere and she got home and she said, “Gina, I hate to tell you this, but you’re going to see a hard braking event on that trip. I’m really sorry, somebody pulled out in front of me.” So I know my story’s anecdotal, but I really enjoy seeing the ability to influence people to be safer on the roads, especially when it comes to keeping my kids safe when family’s driving them.

Louisa Harbage-Edell:

Excellent. Larry, how about you?

Larry Thursby:

I don’t know if you’d call it innovative, but I always found it helpful in conversations about distracted driving to be able to refer to some of the telematics information because that’s one of those conversations that it seemed like you had every day of every year, and no one brought facts, they just brought perceptions there for a while and it was very difficult to try to shut those conversations down until you could inject actual facts. So that was helpful.

Louisa Harbage-Edell:

Absolutely. Teresa?

Teresa Scharn:

I dunno that mine’s innovative. It’s very similar to Gina’s in that the exciting thing for me on telematics is that it can make insurance mean something to customers more than what it does today. Today, people have it – have insurance because they either have to have it, and it’s almost like somewhat of a nuisance to people, to a degree. But with telematics, we have such an opportunity to create that everyday engaging experience. Fun to see my husband who is one of those people who just would never engage with things at all, but when he had that app on, he was watching his driving very, very clearly. And he is not the riskiest driver in the world, but he does make me nervous and my knuckles are white frequently. When I was driving with him, he’s the one who has to pass the car really fast. All of that stopped, and he was watching his driving, he was watching the app, he was engaging with it every day, and we see that. We see that with customers and that they engage. It’s just a tremendous opportunity for us to engage with our customers and have insurance mean something to them, have it be a benefit for them and a benefit for us. And truly, it’s a way for insurance companies to find this type of engagement or to be leading it in telematics, knowing that it’s probably going to extend to other areas in insurance as well.

Louisa Harbage-Edell:

Yeah, absolutely. Robin, how about you?

Robin Harbage:

Louisa mentioned that I’ve worked with telematics for a number of years. In fact, a very early experience was when we took in Western New York and installed 50 cars with hard-installed telematics and with mobile phones. Yes, there was a time when you actually bolted mobile phones into your car. So for those who don’t remember that time it happened, those 50 drivers then had their driving monitored and they had a call-in station where they could call in if they had questions or issues and information was sent back to them. One of our drivers driving around accidentally went off the road, flipped his car, and as he was trying to unbuckle his seatbelt, the call station actually called him up on his hard installed phone and said, “Sir, it looks like your car is parked in unusual position cause of gyroscopic capability.” They knew he was parked upside down. “Would you like us to send out some kind of help to assist you with your situation there?” So that was in the late ’90s. It was innovative at the time. It’s still, in some respects, innovative today because I think first notice of loss and safety features is one of the critical innovations of telematics. And although it’s core to rate making and the primary reason insurers wants to do it, I think for the consumer’s sake and for the industry’s sake, those things that are innovating around safety are key to telematics.

Louisa Harbage-Edell:

Yeah, excellent. So with that behind us, a little bit of background for each of you before we dive into the meat of things, I want to make sure we’re all speaking from a common understanding. So let’s lay a little groundwork and talk a little bit about some of the history of telematics and how it’s evolved for us. So Robin, we’ve obviously established that you’ve been doing this for a while. Can you talk a little bit about some of what we’ve seen to date and how telematics has evolved?

Robin Harbage:

Sure. I think one of the keys here is the technology because when we first started doing telematics, as I said, we use hard installed devices both in Buffalo, New York and in Houston, Texas at Progressive in too early tests. Those were too expensive to be programs that could go out full stream because of the cost. Then we went to install devices in the OBD. Today, the industry standard is clearly to be using mobile apps or to begin using data that’s already inherent in the vehicle because it comes from an OEM through hard solve devices at the factory. And to me, that train of telematics technology is the train of telematics because as you look at telematics, what’s made it hard to go ahead is because anything you ask a consumer to do outside of insurancey things like just filling out an app and paying their premium is really difficult to get them to do.

Louisa Harbage-Edell:

Absolutely. I mean Larry, you piloted one of the early programs as well. Is there anything you’d add to that?

Larry Thursby:

I agree a hundred percent with Robin. The only thing I think I would add is that since those early days, everybody could always say, okay, “I don’t want to go in now because it’s hardwired” – and that’s difficult for the consumer – or “I don’t want to go into the OBD-II port because there’s a lot of wastage of the devices that get sent out or not plugged in” or “I don’t want to go to the phones yet because they drain batteries.” There’s always this reason not to, and I think as an industry we’ve all had to get used to, there’s probably not a perfect end result. You might say, “Well, data’s going to come out of the car someday” and actually it already is [with connected cars], so it already is happening and that’s the end-all be-all. But I just talked to someone the other day that was working on a webcam or basically a camera feed that would compliment the computer data coming out of car so that you could eliminate those. Gina’s example of “I had a hard brake, but I really needed to have the hard brake.” So there’s always going to be someone out there innovating and coming up with a new solution that says, “Here’s the best way or the best source of data.” But I think the key is, as an industry, I see more and more people embracing the here and now of “I’ve got to get started or I’m already in this. I’ve just got to continue to evolve as things change.”

Louisa Harbage-Edell:

Gina, Teresa, anything you’d add to that?

Teresa Scharn:

The only thing I’d say is that one of the things that’s evolved over time is that we used to get data from one or two sources and now we’re getting data from multiple sources. And so the importance of being able to understand and manage data not only at scale but with that level of complexity is something that has grown and is in today’s programs as well.

Gina Minick:

And just last thought on that to Teresa’s point, that really involves all of us working together, which is why we’re here on a panel today. Because if we think about how do we use all of the history that Robin shared and think about the different ways we can collect the data and the data that’s out there in the world right now for us to harness…how do we all work together to harness that data so we can really make this happen with telematics in the industry and have more than 10% of policies signed up with telematics?

Louisa Harbage-Edell:

That’s maybe a really good segue. We’ve seen some monumental shifts this year in driving behavior in response to the COVID pandemic and companies who do have some of that critical data or who’ve been able to access that data have been able to use it to adapt. And Gina, I know you’ve been thinking a lot about that. Do you want to talk a about that a little bit?

Gina Minick:

Yeah, I’ll start with look how Covid has impacted us all. We’re on virtual meetings together. There are four virtual meetings going on in my home right now where I’m at, and consumers are driving less. So we have been tracking actual mileage to understand how are driving trends changing with COVID. That’s really been imposed on all of us across the world and we have seen notable decreases in mileage At times it’s been up to like 50% less mileage now we’re much…in a much more stable place, and miles driven year over year are getting back to a normal place. We’re still seeing people are taking fewer trips. I think, just last week, our data was showing 8% fewer trips were being taken. And we’re watching this really closely to help insurance companies make decisions on how to make rate adjustments in the future and how to keep a pulse on these trends.

But as frustrating as COVID has been for many of us, I like to try to focus on the positive changes I’ve seen. And one very promising trend is that shift in consumer mindset. I’m aware that I’m driving less and I’m aware more and in tune with how I’m driving and it’s a little more top of mind for me to think about telematics and my insurance than it was a year ago today. So Arity actually conducted research back in 2019 asking about consumer’s willingness to share driving data and different pieces of driving data, and we saw that about 30% of consumers were willing to share driving data. But then we ran that same study in May of 2020 and asked the same questions and it was fascinating to see actually there was quite a shift, more like depending on the data element, 40 to 50% willingness to share data. So this is really promising to unprecedented shifts in consumer mindset, which I like to think will open the door for insurance companies to get higher adoption of telematics.

Louisa Harbage-Edell:

You’ve of lived this experience as one of those insurance companies. Do you want to talk at all about some of what you’ve seen this year?

Teresa Scharn:

Yeah, one of the things that the telematics data has enabled really was what Gina was talking to at the beginning, and that is with Covid having an immediate understanding of the changes in mileage and the changes in how people drive and the insurance company’s reaction to that and the speed of that reaction is unprecedented. It would’ve taken us years or year at least, I would say, for those trends to be seen in our data. And if the trends came back, as Gina said, we likely would not have recognized it because trends are back to normal and you’re writing rates for the future, you’re creating rates now for the future. Well, with telematics data, we were able to see immediately how much mileage was dropping, understand exactly what that was doing to our frequencies and calculate that impact on our premiums and give those refunds back to our customers.

And many large insurance companies, many insurance companies did that within months. That is just amazing. And that was 100% enabled by telematics. Without that, nobody would’ve been comfortable doing that and we would not have had the data to do it correctly and profitably as well. So that is very, very exciting. And on the consumer interest side, we are absolutely seeing an increase in interest. We rolled out our SmartMiles, which is a pay-per-mile program last year, and we’re scaling it this year. Sometimes timing is great. And so we have seen a significant increase in the sale of that of all programs of particularly that program, it’s already 10% of our new business, which is significantly higher than we had ever anticipated, a true per-mile program being…and it’s close to 15 million a premium already this year. So exciting to see that.

Louisa Harbage-Edell:

Interesting. Excellent. Alright, so we’ve grown a lot over the past decades. This data is incredibly valuable, but we’ve also acknowledged that we do still see limitations to these programs. That there are still challenges to getting people into them to get the meaningful data necessary to get these kinds of insights. So let’s ask our audience. We have a poll question for you here around what do you see as the biggest challenge currently facing insurance companies to bring these telematics programs to the next level? Can we bring that one up? All right, so what challenges are you seeing as being significant? Is it things around distribution, customer participation, analytics, sophistication, regulation, just generally not having programs in place? We’ll just give this a minute for you all to get your answers in. It looks like those are leveling off. So I think we can go ahead and close out the poll and share the results with you.

So far and away, more than half of you feel like customer participation is the number one issue here. A few things start to filter out sort of analytics sophistication, looks like it’s coming in seconds, distribution and regulation close behind and then just not having a program in place at all certainly is still in there as an issue as well. So customer participation, I mean that’s a big one and we’ve talked a lot about this over time. I mean, I don’t know if any of you have some strong feelings about where you want to start or tackling that customer participation issue in particular.

Robin Harbage:

I can speak to that a bit. It goes to the comment I made earlier about asking customers to do something that’s not insurancey and I think it can’t be overstated that you want to make this as simple as possible for consumers. And I think one thing the audience here has to take away from this is that we’re going there, and it’s got to be a question of how fast you want to move towards this as a insurer because there is so much ambient data that is now available either through the car or through the phone, and there are people that are pulling that data together and going to be making it ready for you to use to work with your consumers. So to the extent that you can do that, so the consumer can participate simply by saying, yes, use my data and the data’s already embedded in the insurer’s capability to pull and look at it, will really change the way that this rolls out.

Larry Thursby:

I think I would add to that, I was one of the folks early on that made the mistake of thinking, consumer interest would be high for this, but for getting that insurance in general, consumer interest is almost next to nothing. So you’ve got something that’s very interesting within a category that’s of no interest at all to 99.9% of the population. So now that I think we understand that, going back to Robin’s thought, this thing is, and something that Gina shared earlier is consumers are not averse to the concept of telematics in general. So if we can embed it within the process where they don’t have to take additional steps that are required today, I think that will make all the difference in the world and we’ll probably take that poll result to 57% consumer interest and show up for what it really is, right. Consumers, I think, will be on board as soon as it’s almost automatic to be part of the process. And I don’t think you’ll see any type of rebellion against the use of telematics from consumers, from consumer advocates or really from regulators. Certainly there are going to be questions out there, but nothing compared to what the insurance industry has gone through with other variables such as use of credit or financial responsibility and rating.

Teresa Scharn:

I think this is a case where when you ask consumers if they’re interested or you put a value prop in front of them, sometimes you get two different responses as well. And so consumers absolutely are and should be concerned about where they share their data and how they’re sharing their data. And as insurance companies, customers don’t trust us. We’re not an area where people trust. And so you have to figure out a way to get past that either by making sure you’re showing them what you’re sharing or what data they’re sharing in a very clear and concise way or by really truly having a value prop in the end when all the consumer studies we’ve done, there’s always that person who says, “I’m never going to share my data.” This isn’t going to happen. And then as soon as you say you can save up to 40% on insurance, and by the way the average is 20, I’ve seen the cybersecurity person pull up his chair and go, “What? Okay, now I’m interested.” And so it’s really in how you present it. and we changed on our front end for the direct channel that goes directly to consumers. We just changed this year how we sold it. We put little videos in explaining it to customers, did it in many different ways. And our consumer adoption went from about 60% to 80, 85% with telematics. So it really is making it simple and meaningful to them in a way that engages them. Yeah.

Gina Minick:

Larry, something you said really resonated with me around, you overestimated consumer’s interest. I think all of us have been in the industry for our careers. So when Teresa, I think we’ve all designed telematics programs and we think, “Yeah, everyone’s going to love this. I love tracking my family, I love looking at this data” – but we’re talking to ourselves. How do we take ourselves out of it and our passion for data and put ourselves in the consumer’s shoes, in the agent’s shoes or just that online flow, the call center rep, how do we make it so easy for them to get that telematics data and onboard a customer who let’s face it, may not love telematics as much as we do, right?

Louisa Harbage-Edell:

Yeah, absolutely.

Larry Thursby:

Teresa’s point, this is a story that’s fun to tell with videos or whatever that else it is and it’s a good thing to talk about. So we don’t necessarily have to shy away with it from it as we have some other things. We’re just not always that great as an industry talking to consumers about stuff like this. We’re great at handling claims, that’s what we’ve done for whatever, how many hundred years we’ve been in auto insurance. But now this is really a different thing for us to talk about. So it seems natural that it would take us a little time

Louisa Harbage-Edell:

And definitely a piece I think that requires some education. So how we do that is important. So anything else you’d add to that consumer participation or that engagement piece before we move on to one of these other issues,

Larry Thursby:

I’m going to ask Robin a question or Teresa or Gina. Robin, you mentioned making it as easy as possible for the consumer to participate in telematics. And so in my mind I go to point of sale data that’s coming from an OEM, right? So the computer’s already in the car, I’m here selling new business. I don’t need to ask you for anything other than your agreement to pull that data in and provide you to Teresa’s point, perhaps a 20%, perhaps a 40%, maybe even larger discount. Where are we at in terms of how long do you think before that’s realistic for insurance companies to use widely?

Robin Harbage:

So the issue of course with vehicles is the fact that even though models have it out there, the average age of a vehicle being 10 years older more means there’s a lot of unequipped vehicles out there and the rate of change in those vehicles slows down innovation. I’ve long said that having a car not just be a vehicle for taking in information, but the vehicle for passing information through as a modem essentially for your phone is the way to get information faster out there because then the innovation comes from the phone which goes out rather quickly. So as I think about this, I think yes, the cars will be there, they will probably be the future of this because it’ll be so rich data and it’ll be available. But in the meantime, over the next two years, phones are going to be clearly the vehicle by which we take data in and the fact that there are so many location derived apps out there on your phone already, all of which are taking a location of data and that data can be pulled together.

That’s where we’re going to see the immediate impact on telematics being adopted of eight people. I guess I’d say, and I strongly emphasize when you look at telematics or why it hasn’t taken place, it has to taken place in many respects because companies haven’t leaned into it in the hard way that a few large insurers have. And by that I mean their senior management say, yeah, we have to do it, but they’re not fully committed to it. If there’s one thing I think has prevented it, the lack of senior management buy-in and commitment to it. Because once they do and they take you to the point where it’s not just an option, it is the option, it is your default for selling insurance, that’s when you really start to push it out there. And you’ll buy a lot of, you’ll find a lot of very anxious customers who are willing to take those discounts and better rates when you put it out there if you make it the default.

Larry Thursby:

I think it’s a great point. One of the things, Robin, I think we talked about a little bit before too, the buy-in. It feels like the buy-in from executives at insurance companies for telematics is very high right now. At least what it was in the past. Everybody sees the future and what needs to happen. What I think it feels like we have now is buy-in to do it, but not a reason. Why are you doing it now? I used this analogy the other day, it’s almost like the mask wearing, right? If you watch sports, and I’m a big sports fan, you can look at the sidelines and see the coach that wears the mask down over their chin. I call it the chin diaper and then the coach that wears it over the mouth. So there’s agreement, okay, I’ve got to wear the mask, but why am I doing it? What’s the point and am I getting at it? And I think that’s the buy-in piece that we still lack a bit, and that’s up to a lot of us to continue to stay in ways to leadership throughout the organization. Here’s why we need to be doing it now and what we get out of it now versus “Yeah, I know what we get out of it in seven years, but why am I investing, whatever it is, $500,000, a million or two million this year in telematics?”

Louisa Harbage-Edell:

And interestingly from our audience that internal buy-in was one of the lowest points of concern. So I think we really have seen and increase in that buy-in and a shift in that area. So the second one that really came up for people was that distribution issue. So Teresa, I know you’ve put a lot of thought into this, you want to talk a little bit about maybe how you’ve overcome some of those distribution challenges.

Teresa Scharn:

And I think I’d say we’re all still overcoming some of those distribution challenges. So we’ve made a lot of headway. So we are largely independent agents or solely independent agents for the most part. We do have a direct channel now and it is very challenging and some things that honestly we could do better. And the same message goes for independent agents as what we just said for customers. You have to make it simple. This is new, this is different, this is changing the way they sell insurance. And that is when it’s confusing and hard, that’s when it just doesn’t get done. And so even having multiple programs, having multiple versions of programs, having multiple…having our programs side by side with a Progressive program right inside of an agency – that’s confusing for agents. And I’d say the nut that we’re trying to crack and still trying to crack is really how do we keep it as simple as possible at point of sale for agents and make it really easy for them to work through the process?

The piece that I think we have done well is helping them understand some of the why and what’s in it for them, which is why they’re selling it. So the 10% participation discount alone from a savings perspective is huge. And then we also have an option where they can get the full participation discount or the full telematics discount if they have data that comes with them so that the competitive pricing upfront, alone helps them tremendously from a sales perspective and then the retention side, understanding that they spend a lot of time to get business in the door, they want to keep it, they don’t want to lose that customer. And so the impacts that we have on retention and making sure helping them understand that I think we’ve gotten past that. That message has gotten out fairly well, but we have room to grow in making it simple.

Louisa Harbage-Edell:

Absolutely. I mean, Robin, Gina, you’ve helped roll out a number of programs and I know that’s been a key concern. Anything you’d add in your thinking there?

Gina Minick:

I think of you threw up many challenges in that question, Louisa said, I think of it as like a progression or peeling back the onion. At the end of the day, the insurance industry wants this data because we want to understand how to price people accurately. Insurance is about understanding risks. So we want the data to understand risk, but we have to get the data by creating an app, creating an engaging experience and getting that consumer data. But to Teresa’s point, we can’t get that data unless we have effective distribution and we make it so easy. So as we keep peeling it back, we think, well, we know what we want problem trying to solve, but it has to start with a simple program in a simple way for agents or again, your online flow to just seamlessly have somebody push through and got to keep in mind insurance companies are trying to bind policies too. So it has to be simple so that you get from quotes to bind and then through to actually get that telematics data so we can get what we need at the end of the day, which is access to that data to better understand the risk.

Louisa Harbage-Edell:

So I think the last of the top three was really around this idea of analytics sophistication and how we think about that, especially when you’ve got this kind of data and all of these variables. Robin, I mean you’ve obviously put a lot of thought into some of these programs and getting to that point. Anything you’d throw on the table with regards to that one?

Robin Harbage:

Certainly. So the sophistication has largely been hamstrung by the fact that all insurers have used this as a discount tacked on to the rest of their rating variables. So if you really want to make this something of import and to use as full power, and let’s be honest, it has the most power of any rating variable that we’ve discovered so far, there’s no question about that. So if you want to make use of that and in fact reduce the reliance on some of the proxies that have some problems, if you will, societally as well as from the standpoint of the insurer’s acceptance of some of those variables like credit, you’d have to say, look, we’re going to make this a part of our generalized linear model. We’re going to just put it in as its full power and let it take off all the handcuffs that we’ve been on it because we used it only on the residual of signal that’s not identified for other proxy variables.

Once you do that and you make it core to your rating, its power will be tremendous. It’ll attract all the best drivers. It’ll segment that company’s data better than anybody else. It’ll be in a kind of innovation that has largely driven some major insurers’ growth for years and years. I’d be surprised if some insurers aren’t within a year away from making that a core to their product and will be gone – the other insurers that don’t have that power once it starts happening. In fact, there are some insurers already out there who have made telematics the core of the entire rating. So I think that largely is where insurers have to go and have to go quickly.

Larry Thursby:

I think the statement you just made, Robin is scary, right? And Nationwide, Teresa has carried this on and done a great job. I think one of the purposes we implemented telematics early on was to get data so that we could do the analysis so that we could be ready when the class plan revolution occurred to make the pivot the same time as some of the true leaders that are out there. And the fact that in the poll there was a large percentage of folks who said analytics sophistication is a concern in terms of its telematics, continued evolution tells you that yes, those carriers who have the data and are ready to begin using telematics as something other than a monovariate discount, take it to the top of your rating equation and vary other things like age and gender and driving points – all those other things begin taking away from those and adding to telematics.

That should be scary to everyone. And I think the question that it raised, at least for me from a Nationwide perspective is, can we as a company go fast enough on our own with our own data, doing our own warehousing, doing our own analytics to build that? Or are we forced to go into with a third party like Arity, like Lexis, some of the other folks who are aggregating even more data so that we are ready to go sooner rather than later? And Teresa, I don’t know if you have any perspective, if you continue to stay the course on we’re going to kind of get all our data, we’re going to do it ourselves or are we going to form that piece out?

Teresa Scharn:

Yeah, and I think we’re always thinking about what’s the most efficient and most effective way to bring this to life. And as we think through that, one of the number one things that I think anybody needs to think about is where do you want to differentiate? Where do you want to differentiate your program and be very clear and be very focused on that. And then why do you want to differentiate on that? And I’ll give you an example to that, going back to this point is that when you think about analytics, there’s lots of steps to this. There’s the data integration, there’s just the data management of all of that data in which you get it in in-house, the data science behind taking that data, normalizing it, making all the data features, and then there’s the modeling on top of it. And this is at a scale that we haven’t seen before.

So which part of each one of those steps is important to be really fast at, which part of these steps is really important to have be Nationwide only, which part isn’t? And I think you can partner with any piece and part of that. Today, Nationwide does do all of that ourselves, and we do a lot of it in-house now. We didn’t start in-house, we started by partnering with a company with Robin actually at the very beginning, right Larry, where we started partnering and used their scoring and their rating, but ended up building it all. But I would say that’s all TBD to bring in data very fast to integrate with new customers faster and new data sources faster. Will we be able to do that better if we partner with somebody versus build it ourselves? That’s something we’re actually asking right now and actually have an RFP out right now to figure out that solution. And it’s because we understand what our goal is, we understand where we want to differentiate and what the end solution needs to be and it might be better to partner to do that.

Larry Thursby:

And just so folks don’t feel like they’re doing something wrong, that’s hard. But I can remember when we started, we had a minimal number of customers in the program and we couldn’t get the data across communication lines into our warehouse. That took us months, then we got it into a warehouse and we couldn’t figure out how to get the data out of the warehouse to actually begin to analyze it and study it. And that was with one source. And now as you’ve said previously, you’re getting data from multiple cell phone, third party providers from OBD-II, from OEMs, and the amount of data is so much larger. That is a task that’s monumental for just about any insurance company, no matter how good they are at data science. So if there’s folks out there struggling with it, they shouldn’t feel like they’re alone. Everybody I think save maybe one or two struggles with that.

Gina Minick:

And I think this does apply to all insurance companies regardless of size. It’s like how do you find the right recipe for you based on where your journey is on telematics, what are the things that you need help from other vendors on because you don’t have the capabilities because you don’t have them yet because you’ll never build ’em because it’s not core to your tech stack and it’s something you don’t need to invest in. How do you balance that level of investment through the whole data flow, all the way

Louisa Harbage-Edell:

“Telematics will revolutionize the insurance industry.” How many of you have heard this bold statement, and how many of you are still waiting for that meaningful change to really come to fruition five, 10, or even 15 years later? My name is Louisa Harbage-Edell, and today I’m joined by four industry experts for this bootcamp on innovating the future of auto insurance. We’ve reached a point where a large number of companies have a telematics program in place, but many still struggle to gain critical mass. In fact, less than 10% of policies enforced today across the industry are usage-based auto insurance ones. So what’s holding us back, and how can we get past these barriers to achieve the opportunities we see for telematics in the future? In today’s informal discussion, my panelists will address these issues and more. We’ll also be asking you, our audience, some poll questions to get your thoughts, and we’ll open it up for your questions via live Q&A towards the end of the session. That said, you’ll see that Q&A panel is open throughout, so feel free to drop in your questions at any time. By the end of this conversation, you’ll have ideas for how to overcome some of the key challenges still facing companies today in order to take your program to the next level, as well as key insight into the exciting new ways telematics is evolving for the future. With that, let me introduce our panelists.

First up, Gina Minick is product director for Insurance Solutions at Arity. She has extensive experience in product management as well as, prior to Arity, being at Allstate, where she played an integral role in the countrywide rollout of Drivewise, Allstate’s telematics program. She’s also served in numerous other roles at Allstate and Progressive, including product strategy and implementation, risk management, catastrophe management, competitive intelligence and marketing. Larry Thursby is a P&C insurance consultant and a former Nationwide executive. He worked at Nationwide for 25 years before retiring in 2017. His final 10 years at the company were as vice president of personalized product and pricing, and it was of course during this time that Nationwide introduced their first UBI program for consumers. After initial success in that pilot state, that offering grew into the countrywide program including OBD-II [On-Board Diagnostic II] and smartphone connections as well as connected car solutions.

Teresa Scharn is the current VP for personal lines product development at Nationwide and has worked in the P&C insurance industry for over 25 years with over 20 of those at Nationwide. She has experience in pricing, underwriting strategy, product management, and product development, and currently leads the team responsible for new and innovative products for all personal lines programs for Nationwide’s, industry leading SmartRide and SmartMiles UBI programs. Robin Harbage is a telematics consultant and former insurance industry executive. Robin has over 40 years of experience in roles such as chief actuary, insurance, product manager, and general manager overseeing pricing, marketing, and product development for top tier insurance carriers, most notably Progressive. And in addition, spent over a decade in consulting where he built and brought the drivability program to market and helped dozens of insurance companies launch their initial telematics offerings. He has, in fact, worked with telematics products since 1996. So let’s get to know you all a little bit better. I would love to hear from each of you some interesting innovation or application of telematics you’ve all experienced during your time working with these different products. So Gina, maybe you could kick us off.

Gina Minick:

Sure. Thanks, Louisa. I’ll share a more personal story. Louisa mentioned I’ve shared, I’ve spent my entire career in the insurance industry – about half of that in telematics. And at this point my entire family knows that I believe very strongly in safe driving and they welcome, I think, they welcome that I impose my beliefs on them by connecting all of the things in our lives, including those of my extended family. So my most enjoyable use of telematics that I’ve seen was once when my mother-in-law actually borrowed my car to drive my kids somewhere and she got home and she said, “Gina, I hate to tell you this, but you’re going to see a hard braking event on that trip. I’m really sorry, somebody pulled out in front of me.” So I know my story’s anecdotal, but I really enjoy seeing the ability to influence people to be safer on the roads, especially when it comes to keeping my kids safe when family’s driving them.

Louisa Harbage-Edell:

Excellent. Larry, how about you?

Larry Thursby:

I don’t know if you’d call it innovative, but I always found it helpful in conversations about distracted driving to be able to refer to some of the telematics information because that’s one of those conversations that it seemed like you had every day of every year, and no one brought facts, they just brought perceptions there for a while and it was very difficult to try to shut those conversations down until you could inject actual facts. So that was helpful.

Louisa Harbage-Edell:

Absolutely. Teresa?

Teresa Scharn:

I dunno that mine’s innovative. It’s very similar to Gina’s in that the exciting thing for me on telematics is that it can make insurance mean something to customers more than what it does today. Today, people have it – have insurance because they either have to have it, and it’s almost like somewhat of a nuisance to people, to a degree. But with telematics, we have such an opportunity to create that everyday engaging experience. Fun to see my husband who is one of those people who just would never engage with things at all, but when he had that app on, he was watching his driving very, very clearly. And he is not the riskiest driver in the world, but he does make me nervous and my knuckles are white frequently. When I was driving with him, he’s the one who has to pass the car really fast. All of that stopped, and he was watching his driving, he was watching the app, he was engaging with it every day, and we see that. We see that with customers and that they engage. It’s just a tremendous opportunity for us to engage with our customers and have insurance mean something to them, have it be a benefit for them and a benefit for us. And truly, it’s a way for insurance companies to find this type of engagement or to be leading it in telematics, knowing that it’s probably going to extend to other areas in insurance as well.

Louisa Harbage-Edell:

Yeah, absolutely. Robin, how about you?

Robin Harbage:

Louisa mentioned that I’ve worked with telematics for a number of years. In fact, a very early experience was when we took in Western New York and installed 50 cars with hard-installed telematics and with mobile phones. Yes, there was a time when you actually bolted mobile phones into your car. So for those who don’t remember that time it happened, those 50 drivers then had their driving monitored and they had a call-in station where they could call in if they had questions or issues and information was sent back to them. One of our drivers driving around accidentally went off the road, flipped his car, and as he was trying to unbuckle his seatbelt, the call station actually called him up on his hard installed phone and said, “Sir, it looks like your car is parked in unusual position cause of gyroscopic capability.” They knew he was parked upside down. “Would you like us to send out some kind of help to assist you with your situation there?” So that was in the late ’90s. It was innovative at the time. It’s still, in some respects, innovative today because I think first notice of loss and safety features is one of the critical innovations of telematics. And although it’s core to rate making and the primary reason insurers wants to do it, I think for the consumer’s sake and for the industry’s sake, those things that are innovating around safety are key to telematics.

Louisa Harbage-Edell:

Yeah, excellent. So with that behind us, a little bit of background for each of you before we dive into the meat of things, I want to make sure we’re all speaking from a common understanding. So let’s lay a little groundwork and talk a little bit about some of the history of telematics and how it’s evolved for us. So Robin, we’ve obviously established that you’ve been doing this for a while. Can you talk a little bit about some of what we’ve seen to date and how telematics has evolved?

Robin Harbage:

Sure. I think one of the keys here is the technology because when we first started doing telematics, as I said, we use hard installed devices both in Buffalo, New York and in Houston, Texas at Progressive in too early tests. Those were too expensive to be programs that could go out full stream because of the cost. Then we went to install devices in the OBD. Today, the industry standard is clearly to be using mobile apps or to begin using data that’s already inherent in the vehicle because it comes from an OEM through hard solve devices at the factory. And to me, that train of telematics technology is the train of telematics because as you look at telematics, what’s made it hard to go ahead is because anything you ask a consumer to do outside of insurancey things like just filling out an app and paying their premium is really difficult to get them to do.

Louisa Harbage-Edell:

Absolutely. I mean Larry, you piloted one of the early programs as well. Is there anything you’d add to that?

Larry Thursby:

I agree a hundred percent with Robin. The only thing I think I would add is that since those early days, everybody could always say, okay, “I don’t want to go in now because it’s hardwired” – and that’s difficult for the consumer – or “I don’t want to go into the OBD-II port because there’s a lot of wastage of the devices that get sent out or not plugged in” or “I don’t want to go to the phones yet because they drain batteries.” There’s always this reason not to, and I think as an industry we’ve all had to get used to, there’s probably not a perfect end result. You might say, “Well, data’s going to come out of the car someday” and actually it already is [with connected cars], so it already is happening and that’s the end-all be-all. But I just talked to someone the other day that was working on a webcam or basically a camera feed that would compliment the computer data coming out of car so that you could eliminate those. Gina’s example of “I had a hard brake, but I really needed to have the hard brake.” So there’s always going to be someone out there innovating and coming up with a new solution that says, “Here’s the best way or the best source of data.” But I think the key is, as an industry, I see more and more people embracing the here and now of “I’ve got to get started or I’m already in this. I’ve just got to continue to evolve as things change.”

Louisa Harbage-Edell:

Gina, Teresa, anything you’d add to that?

Teresa Scharn:

The only thing I’d say is that one of the things that’s evolved over time is that we used to get data from one or two sources and now we’re getting data from multiple sources. And so the importance of being able to understand and manage data not only at scale but with that level of complexity is something that has grown and is in today’s programs as well.

Gina Minick:

And just last thought on that to Teresa’s point, that really involves all of us working together, which is why we’re here on a panel today. Because if we think about how do we use all of the history that Robin shared and think about the different ways we can collect the data and the data that’s out there in the world right now for us to harness…how do we all work together to harness that data so we can really make this happen with telematics in the industry and have more than 10% of policies signed up with telematics?

Louisa Harbage-Edell:

That’s maybe a really good segue. We’ve seen some monumental shifts this year in driving behavior in response to the COVID pandemic and companies who do have some of that critical data or who’ve been able to access that data have been able to use it to adapt. And Gina, I know you’ve been thinking a lot about that. Do you want to talk a about that a little bit?

Gina Minick:

Yeah, I’ll start with look how Covid has impacted us all. We’re on virtual meetings together. There are four virtual meetings going on in my home right now where I’m at, and consumers are driving less. So we have been tracking actual mileage to understand how are driving trends changing with COVID. That’s really been imposed on all of us across the world and we have seen notable decreases in mileage At times it’s been up to like 50% less mileage now we’re much…in a much more stable place, and miles driven year over year are getting back to a normal place. We’re still seeing people are taking fewer trips. I think, just last week, our data was showing 8% fewer trips were being taken. And we’re watching this really closely to help insurance companies make decisions on how to make rate adjustments in the future and how to keep a pulse on these trends.

But as frustrating as COVID has been for many of us, I like to try to focus on the positive changes I’ve seen. And one very promising trend is that shift in consumer mindset. I’m aware that I’m driving less and I’m aware more and in tune with how I’m driving and it’s a little more top of mind for me to think about telematics and my insurance than it was a year ago today. So Arity actually conducted research back in 2019 asking about consumer’s willingness to share driving data and different pieces of driving data, and we saw that about 30% of consumers were willing to share driving data. But then we ran that same study in May of 2020 and asked the same questions and it was fascinating to see actually there was quite a shift, more like depending on the data element, 40 to 50% willingness to share data. So this is really promising to unprecedented shifts in consumer mindset, which I like to think will open the door for insurance companies to get higher adoption of telematics.

Louisa Harbage-Edell:

You’ve of lived this experience as one of those insurance companies. Do you want to talk at all about some of what you’ve seen this year?

Teresa Scharn:

Yeah, one of the things that the telematics data has enabled really was what Gina was talking to at the beginning, and that is with Covid having an immediate understanding of the changes in mileage and the changes in how people drive and the insurance company’s reaction to that and the speed of that reaction is unprecedented. It would’ve taken us years or year at least, I would say, for those trends to be seen in our data. And if the trends came back, as Gina said, we likely would not have recognized it because trends are back to normal and you’re writing rates for the future, you’re creating rates now for the future. Well, with telematics data, we were able to see immediately how much mileage was dropping, understand exactly what that was doing to our frequencies and calculate that impact on our premiums and give those refunds back to our customers.

And many large insurance companies, many insurance companies did that within months. That is just amazing. And that was 100% enabled by telematics. Without that, nobody would’ve been comfortable doing that and we would not have had the data to do it correctly and profitably as well. So that is very, very exciting. And on the consumer interest side, we are absolutely seeing an increase in interest. We rolled out our SmartMiles, which is a pay-per-mile program last year, and we’re scaling it this year. Sometimes timing is great. And so we have seen a significant increase in the sale of that of all programs of particularly that program, it’s already 10% of our new business, which is significantly higher than we had ever anticipated, a true per-mile program being…and it’s close to 15 million a premium already this year. So exciting to see that.

Louisa Harbage-Edell:

Interesting. Excellent. Alright, so we’ve grown a lot over the past decades. This data is incredibly valuable, but we’ve also acknowledged that we do still see limitations to these programs. That there are still challenges to getting people into them to get the meaningful data necessary to get these kinds of insights. So let’s ask our audience. We have a poll question for you here around what do you see as the biggest challenge currently facing insurance companies to bring these telematics programs to the next level? Can we bring that one up? All right, so what challenges are you seeing as being significant? Is it things around distribution, customer participation, analytics, sophistication, regulation, just generally not having programs in place? We’ll just give this a minute for you all to get your answers in. It looks like those are leveling off. So I think we can go ahead and close out the poll and share the results with you.

So far and away, more than half of you feel like customer participation is the number one issue here. A few things start to filter out sort of analytics sophistication, looks like it’s coming in seconds, distribution and regulation close behind and then just not having a program in place at all certainly is still in there as an issue as well. So customer participation, I mean that’s a big one and we’ve talked a lot about this over time. I mean, I don’t know if any of you have some strong feelings about where you want to start or tackling that customer participation issue in particular.

Robin Harbage:

I can speak to that a bit. It goes to the comment I made earlier about asking customers to do something that’s not insurancey and I think it can’t be overstated that you want to make this as simple as possible for consumers. And I think one thing the audience here has to take away from this is that we’re going there, and it’s got to be a question of how fast you want to move towards this as a insurer because there is so much ambient data that is now available either through the car or through the phone, and there are people that are pulling that data together and going to be making it ready for you to use to work with your consumers. So to the extent that you can do that, so the consumer can participate simply by saying, yes, use my data and the data’s already embedded in the insurer’s capability to pull and look at it, will really change the way that this rolls out.

Larry Thursby:

I think I would add to that, I was one of the folks early on that made the mistake of thinking, consumer interest would be high for this, but for getting that insurance in general, consumer interest is almost next to nothing. So you’ve got something that’s very interesting within a category that’s of no interest at all to 99.9% of the population. So now that I think we understand that, going back to Robin’s thought, this thing is, and something that Gina shared earlier is consumers are not averse to the concept of telematics in general. So if we can embed it within the process where they don’t have to take additional steps that are required today, I think that will make all the difference in the world and we’ll probably take that poll result to 57% consumer interest and show up for what it really is, right. Consumers, I think, will be on board as soon as it’s almost automatic to be part of the process. And I don’t think you’ll see any type of rebellion against the use of telematics from consumers, from consumer advocates or really from regulators. Certainly there are going to be questions out there, but nothing compared to what the insurance industry has gone through with other variables such as use of credit or financial responsibility and rating.

Teresa Scharn:

I think this is a case where when you ask consumers if they’re interested or you put a value prop in front of them, sometimes you get two different responses as well. And so consumers absolutely are and should be concerned about where they share their data and how they’re sharing their data. And as insurance companies, customers don’t trust us. We’re not an area where people trust. And so you have to figure out a way to get past that either by making sure you’re showing them what you’re sharing or what data they’re sharing in a very clear and concise way or by really truly having a value prop in the end when all the consumer studies we’ve done, there’s always that person who says, “I’m never going to share my data.” This isn’t going to happen. And then as soon as you say you can save up to 40% on insurance, and by the way the average is 20, I’ve seen the cybersecurity person pull up his chair and go, “What? Okay, now I’m interested.” And so it’s really in how you present it. and we changed on our front end for the direct channel that goes directly to consumers. We just changed this year how we sold it. We put little videos in explaining it to customers, did it in many different ways. And our consumer adoption went from about 60% to 80, 85% with telematics. So it really is making it simple and meaningful to them in a way that engages them. Yeah.

Gina Minick:

Larry, something you said really resonated with me around, you overestimated consumer’s interest. I think all of us have been in the industry for our careers. So when Teresa, I think we’ve all designed telematics programs and we think, “Yeah, everyone’s going to love this. I love tracking my family, I love looking at this data” – but we’re talking to ourselves. How do we take ourselves out of it and our passion for data and put ourselves in the consumer’s shoes, in the agent’s shoes or just that online flow, the call center rep, how do we make it so easy for them to get that telematics data and onboard a customer who let’s face it, may not love telematics as much as we do, right?

Louisa Harbage-Edell:

Yeah, absolutely.

Larry Thursby:

Teresa’s point, this is a story that’s fun to tell with videos or whatever that else it is and it’s a good thing to talk about. So we don’t necessarily have to shy away with it from it as we have some other things. We’re just not always that great as an industry talking to consumers about stuff like this. We’re great at handling claims, that’s what we’ve done for whatever, how many hundred years we’ve been in auto insurance. But now this is really a different thing for us to talk about. So it seems natural that it would take us a little time

Louisa Harbage-Edell:

And definitely a piece I think that requires some education. So how we do that is important. So anything else you’d add to that consumer participation or that engagement piece before we move on to one of these other issues,

Larry Thursby:

I’m going to ask Robin a question or Teresa or Gina. Robin, you mentioned making it as easy as possible for the consumer to participate in telematics. And so in my mind I go to point of sale data that’s coming from an OEM, right? So the computer’s already in the car, I’m here selling new business. I don’t need to ask you for anything other than your agreement to pull that data in and provide you to Teresa’s point, perhaps a 20%, perhaps a 40%, maybe even larger discount. Where are we at in terms of how long do you think before that’s realistic for insurance companies to use widely?

Robin Harbage:

So the issue of course with vehicles is the fact that even though models have it out there, the average age of a vehicle being 10 years older more means there’s a lot of unequipped vehicles out there and the rate of change in those vehicles slows down innovation. I’ve long said that having a car not just be a vehicle for taking in information, but the vehicle for passing information through as a modem essentially for your phone is the way to get information faster out there because then the innovation comes from the phone which goes out rather quickly. So as I think about this, I think yes, the cars will be there, they will probably be the future of this because it’ll be so rich data and it’ll be available. But in the meantime, over the next two years, phones are going to be clearly the vehicle by which we take data in and the fact that there are so many location derived apps out there on your phone already, all of which are taking a location of data and that data can be pulled together.

That’s where we’re going to see the immediate impact on telematics being adopted of eight people. I guess I’d say, and I strongly emphasize when you look at telematics or why it hasn’t taken place, it has to taken place in many respects because companies haven’t leaned into it in the hard way that a few large insurers have. And by that I mean their senior management say, yeah, we have to do it, but they’re not fully committed to it. If there’s one thing I think has prevented it, the lack of senior management buy-in and commitment to it. Because once they do and they take you to the point where it’s not just an option, it is the option, it is your default for selling insurance, that’s when you really start to push it out there. And you’ll buy a lot of, you’ll find a lot of very anxious customers who are willing to take those discounts and better rates when you put it out there if you make it the default.

Larry Thursby:

I think it’s a great point. One of the things, Robin, I think we talked about a little bit before too, the buy-in. It feels like the buy-in from executives at insurance companies for telematics is very high right now. At least what it was in the past. Everybody sees the future and what needs to happen. What I think it feels like we have now is buy-in to do it, but not a reason. Why are you doing it now? I used this analogy the other day, it’s almost like the mask wearing, right? If you watch sports, and I’m a big sports fan, you can look at the sidelines and see the coach that wears the mask down over their chin. I call it the chin diaper and then the coach that wears it over the mouth. So there’s agreement, okay, I’ve got to wear the mask, but why am I doing it? What’s the point and am I getting at it? And I think that’s the buy-in piece that we still lack a bit, and that’s up to a lot of us to continue to stay in ways to leadership throughout the organization. Here’s why we need to be doing it now and what we get out of it now versus “Yeah, I know what we get out of it in seven years, but why am I investing, whatever it is, $500,000, a million or two million this year in telematics?”

Louisa Harbage-Edell:

And interestingly from our audience that internal buy-in was one of the lowest points of concern. So I think we really have seen and increase in that buy-in and a shift in that area. So the second one that really came up for people was that distribution issue. So Teresa, I know you’ve put a lot of thought into this, you want to talk a little bit about maybe how you’ve overcome some of those distribution challenges.

Teresa Scharn:

And I think I’d say we’re all still overcoming some of those distribution challenges. So we’ve made a lot of headway. So we are largely independent agents or solely independent agents for the most part. We do have a direct channel now and it is very challenging and some things that honestly we could do better. And the same message goes for independent agents as what we just said for customers. You have to make it simple. This is new, this is different, this is changing the way they sell insurance. And that is when it’s confusing and hard, that’s when it just doesn’t get done. And so even having multiple programs, having multiple versions of programs, having multiple…having our programs side by side with a Progressive program right inside of an agency – that’s confusing for agents. And I’d say the nut that we’re trying to crack and still trying to crack is really how do we keep it as simple as possible at point of sale for agents and make it really easy for them to work through the process?

The piece that I think we have done well is helping them understand some of the why and what’s in it for them, which is why they’re selling it. So the 10% participation discount alone from a savings perspective is huge. And then we also have an option where they can get the full participation discount or the full telematics discount if they have data that comes with them so that the competitive pricing upfront, alone helps them tremendously from a sales perspective and then the retention side, understanding that they spend a lot of time to get business in the door, they want to keep it, they don’t want to lose that customer. And so the impacts that we have on retention and making sure helping them understand that I think we’ve gotten past that. That message has gotten out fairly well, but we have room to grow in making it simple.

Louisa Harbage-Edell:

Absolutely. I mean, Robin, Gina, you’ve helped roll out a number of programs and I know that’s been a key concern. Anything you’d add in your thinking there?

Gina Minick:

I think of you threw up many challenges in that question, Louisa said, I think of it as like a progression or peeling back the onion. At the end of the day, the insurance industry wants this data because we want to understand how to price people accurately. Insurance is about understanding risks. So we want the data to understand risk, but we have to get the data by creating an app, creating an engaging experience and getting that consumer data. But to Teresa’s point, we can’t get that data unless we have effective distribution and we make it so easy. So as we keep peeling it back, we think, well, we know what we want problem trying to solve, but it has to start with a simple program in a simple way for agents or again, your online flow to just seamlessly have somebody push through and got to keep in mind insurance companies are trying to bind policies too. So it has to be simple so that you get from quotes to bind and then through to actually get that telematics data so we can get what we need at the end of the day, which is access to that data to better understand the risk.

Louisa Harbage-Edell:

So I think the last of the top three was really around this idea of analytics sophistication and how we think about that, especially when you’ve got this kind of data and all of these variables. Robin, I mean you’ve obviously put a lot of thought into some of these programs and getting to that point. Anything you’d throw on the table with regards to that one?

Robin Harbage:

Certainly. So the sophistication has largely been hamstrung by the fact that all insurers have used this as a discount tacked on to the rest of their rating variables. So if you really want to make this something of import and to use as full power, and let’s be honest, it has the most power of any rating variable that we’ve discovered so far, there’s no question about that. So if you want to make use of that and in fact reduce the reliance on some of the proxies that have some problems, if you will, societally as well as from the standpoint of the insurer’s acceptance of some of those variables like credit, you’d have to say, look, we’re going to make this a part of our generalized linear model. We’re going to just put it in as its full power and let it take off all the handcuffs that we’ve been on it because we used it only on the residual of signal that’s not identified for other proxy variables.

Once you do that and you make it core to your rating, its power will be tremendous. It’ll attract all the best drivers. It’ll segment that company’s data better than anybody else. It’ll be in a kind of innovation that has largely driven some major insurers’ growth for years and years. I’d be surprised if some insurers aren’t within a year away from making that a core to their product and will be gone – the other insurers that don’t have that power once it starts happening. In fact, there are some insurers already out there who have made telematics the core of the entire rating. So I think that largely is where insurers have to go and have to go quickly.

Larry Thursby:

I think the statement you just made, Robin is scary, right? And Nationwide, Teresa has carried this on and done a great job. I think one of the purposes we implemented telematics early on was to get data so that we could do the analysis so that we could be ready when the class plan revolution occurred to make the pivot the same time as some of the true leaders that are out there. And the fact that in the poll there was a large percentage of folks who said analytics sophistication is a concern in terms of its telematics, continued evolution tells you that yes, those carriers who have the data and are ready to begin using telematics as something other than a monovariate discount, take it to the top of your rating equation and vary other things like age and gender and driving points – all those other things begin taking away from those and adding to telematics.

That should be scary to everyone. And I think the question that it raised, at least for me from a Nationwide perspective is, can we as a company go fast enough on our own with our own data, doing our own warehousing, doing our own analytics to build that? Or are we forced to go into with a third party like Arity, like Lexis, some of the other folks who are aggregating even more data so that we are ready to go sooner rather than later? And Teresa, I don’t know if you have any perspective, if you continue to stay the course on we’re going to kind of get all our data, we’re going to do it ourselves or are we going to form that piece out?

Teresa Scharn:

Yeah, and I think we’re always thinking about what’s the most efficient and most effective way to bring this to life. And as we think through that, one of the number one things that I think anybody needs to think about is where do you want to differentiate? Where do you want to differentiate your program and be very clear and be very focused on that. And then why do you want to differentiate on that? And I’ll give you an example to that, going back to this point is that when you think about analytics, there’s lots of steps to this. There’s the data integration, there’s just the data management of all of that data in which you get it in in-house, the data science behind taking that data, normalizing it, making all the data features, and then there’s the modeling on top of it. And this is at a scale that we haven’t seen before.

So which part of each one of those steps is important to be really fast at, which part of these steps is really important to have be Nationwide only, which part isn’t? And I think you can partner with any piece and part of that. Today, Nationwide does do all of that ourselves, and we do a lot of it in-house now. We didn’t start in-house, we started by partnering with a company with Robin actually at the very beginning, right Larry, where we started partnering and used their scoring and their rating, but ended up building it all. But I would say that’s all TBD to bring in data very fast to integrate with new customers faster and new data sources faster. Will we be able to do that better if we partner with somebody versus build it ourselves? That’s something we’re actually asking right now and actually have an RFP out right now to figure out that solution. And it’s because we understand what our goal is, we understand where we want to differentiate and what the end solution needs to be and it might be better to partner to do that.

Larry Thursby:

And just so folks don’t feel like they’re doing something wrong, that’s hard. But I can remember when we started, we had a minimal number of customers in the program and we couldn’t get the data across communication lines into our warehouse. That took us months, then we got it into a warehouse and we couldn’t figure out how to get the data out of the warehouse to actually begin to analyze it and study it. And that was with one source. And now as you’ve said previously, you’re getting data from multiple cell phone, third party providers from OBD-II, from OEMs, and the amount of data is so much larger. That is a task that’s monumental for just about any insurance company, no matter how good they are at data science. So if there’s folks out there struggling with it, they shouldn’t feel like they’re alone. Everybody I think save maybe one or two struggles with that.

Gina Minick:

And I think this does apply to all insurance companies regardless of size. It’s like how do you find the right recipe for you based on where your journey is on telematics, what are the things that you need help from other vendors on because you don’t have the capabilities because you don’t have them yet because you’ll never build ’em because it’s not core to your tech stack and it’s something you don’t need to invest in. How do you balance that level of investment through the whole data flow, all the way from getting the data in to deriving the analytics on the backend?

I think to be successful in this, you cannot be afraid to partner in many different areas. I think our partnerships have been critical to getting us to where we’re at today and finding the right partner, understanding what you want to gain from that experience and working on it together. You’ll grow your knowledge and your learning and your program faster.

Robin Harbage:

I think the issue of the many demanding, which we’ve discussed many times in circles like this is one that people have to think about. There are so many places where data can come from and will come from, and there’s so many insurers out there wanting to get that data. So as an industry, we have to be willing to think about creating portals, whether it’s a Lexis, whomever it might be, where they can collect data from. Many sources overcome the problem that Gina alluded to earlier, which is to try and standardize that data to mend it together into a sensible conglomerate data that can be used then effectively and then pass it because there can’t be more than even a half dozen insurers that have the scale and capability to create links to more than two or three or four different sources of data. And let’s be honest, we’re probably going to see 50, 60, a hundred different sources of data within the next couple of years where people are going to pull from in order to get a truly ubiquitous set of data they’re going to use for telematics.

Louisa Harbage-Edell:

So maybe that’s a good pivot to what we see is coming in the future for telematics. And actually we have a poll question for our audience here around what you see as some of the exciting opportunities for telematics in the future. So if we can pull that up. Do you see the opportunity to replace traditional variables, use telematics for targeted advertising, use telematics to price at point of sale? Because of course right now, realistically we’re using it at point of renewal once you’ve had a chance to actually collect data or something else maybe we haven’t even envisioned and included in this poll here.

Larry Thursby:

Speaking of analytics, it looks like we can add up to more than a hundred percent.

Louisa Harbage-Edell:

I think we allowed people to answer with more than one for this question. Okay.

[Offscreen]:

I was going to say all the above. Yeah.

Louisa Harbage-Edell:

Excellent. All right, so if we close this out and share the results with our panelists, it looks like it’s settled out and it looks like the two big winners here sort of ran away, were replacing traditional variables and using telematics to price at point of sale, and then some pieces around targeted advertising and some other votes as well. We talked a little bit about replacing traditional variables already. I don’t know, Robin, I know you’ve put a lot of thought around this. Do you want to talk a little bit more around that and most of the things you’re seeing?

Robin Harbage:

Well, obviously there’s a lot of proxy variables out there. There’s actually been a lot of information in the news recently about problems with some of these judicial rating variables in that there’s disparity impact against minorities because of the way that those variables are overrepresented in certain minority groups. I think telematics has a wonderful outcome that because it is behaviorally based, not socioeconomically based, that is a variable that should have strong strength as we go forward. This is me speaking from a personal mindset, something about which I’m passionate and I know those will be difficult, but given as I said earlier, telematics and behavior-based driving data is so strong in segmenting drivers and most people accept the fact that how I drive should relate to how I’m insured. I think it’ll be an easy concept to say I’ll put less weight on some of the proxy variables, especially problematic variables and more weight on this as we move forward.

Louisa Harbage-Edell:

Teresa, I know you’ve put some thought into rating in the long term and some of the shifts that may happen there. Do you want to talk at all about any of that?

Teresa Scharn:

Yeah, just to what Robin said. I think with telematics, we know today that we have a lot of proxy variables out there for how people drive. That could be your driver age, that could be your male or female, your marital status, even whether you got caught speeding or not to a degree, could be about where the cops may be located. And so there’s just a tremendous opportunity for telematics to replace a ton of those variables ideally, but at a minimum to allow us to really rate on how people drive and remove a lot of those proxy variables, that creates a tremendously better customer experience overall because they can more clearly understand why am I getting this rate and how am I influencing this rate? They can’t understand how a lot of the things are influencing the rate, even their age to a degree. How does that really influence my rate?

But I can understand if I drove erratically or if I was speeding, how that can impact your rate. Two, the segmentation value that it adds, the ability to be able to rate more accurately at an individual level, which then helps us become more profitable and retain customers longer is tremendous. And then ideally, if you think about it, we can remove a lot of those proxy variables that we’re paying money for. So from a cost perspective, it potentially has today… we’re getting data for – telematics data – and paying for that, and we’re paying for MVRs and we’re paying for some clue and we’re paying for credit. So there’s a potential opportunity in the future for us to actually have a potentially more cost effective outcome by removing some of those variables and being a better segmented. And then think about that at renewals, sometimes we run MVRs at renewals and all these things at renewals, just the cost of it and the customer experience alone both. So we absolutely believe in this and the value in rating, we’re completely heading towards what Robin said, and we expect to have what we call a telematics-based class plan the future. I’m an actuary, it blows my mind to think about what that’s going to look like, but it’s an exciting time to be an actuary, exciting time to be in insurance, and it’s absolutely going to change how we rate.

Louisa Harbage-Edell:

In fact, Larry, I think in an earlier conversation you made an interesting point about if you were starting an insurance company from scratch today and you had all the available variables out there.

Larry Thursby:

Yeah, exactly. So the way we do telematics today, or at least we did at Nationwide, was we had our traditional class plans started with territory. Teresa shared some other variables, and then we use telematics at the end to kind of explain the residual, the unexplained piece. And it’s still, as you would expect, amazingly powerful just explaining the residual. But imagine if you did start from scratch and just said, I’m going to use what I think is probably my most predictive variable first, right? The behavioral driving of the vehicle, and then I’ll use the other things to describe whatever might be left. I think you’d probably get rid of a ton of stuff. There might still be value there, but it might become so little that why argue, why include gender, which I think is probably not intuitive to very many folks, why it should impact your insurance rates if it only matters one or 2% when you’ve got telematics to do the rest. Now that [inaudible] a thousand percent excited about the future of rating and be able to use it even more broadly in the class plan.

I think [inaudible] too utopian, we also have to take a step back and recognize there’ll be challenges in terms of “What do I do at renewal?” How static is driving behavior, how much do I want to reflect changes? Because another thing we know about insurance, I said earlier, it’s not a high consumer interest category. You probably don’t want your bills jumping up and down month to month based on how you’ve driven. And so there’s going to be a lot of fine-tuning that needs to happen. Again, you go back to the, well, how do we do that? I think this is probably a blind spot for the panelists we’ve got here because we don’t say data. I need to know first, how much does the score vary? Why does it vary? How can I soften that? What are ways that regulators will allow me to do to soften that? What if I do soften it and then another carrier goes out and the score’s actually gotten better and they offer a lower rate. So the good news is there’s no lack of work for actuaries in the future, but it should be exciting to watch telematics take over and become much, much, much more powerful in our rating.

Robin Harbage:

Yes, I would mention for the actuaries among us, they fully understand credibility. Credibility is simply a concept of your data; unless it’s large scale is not fully believable, isn’t fully credible. So with any one person’s behavioral data, if it’s a short window of time for which you collect this data, it may not be stable and may not be fully credible. So some kind of rolling average or expansion of the data over time, because people do tend to move back towards the mean over time as a way of stabilizing. So you wouldn’t jump immediately to, “Oh, this month was different than last month.” Give them a new score. And I think those are the types of methodology things that people will adopt that will help with that kind of fluctuation or rate.

Teresa Scharn:

Well, the other thing that it offers is the ability to show people what is going to happen. They’re trending, their driving is trending, and so at least you could set expectations to a degree with telematics. Whereas I don’t know what the base rates are going to change for insurance six months from now. Today we have some trends and have some of an understanding, but I don’t have control over that. The driver actually can have control of that, and if we have the right level of communication with them, they actually might be more comfortable maybe with rates changing a little bit more at renewal. Who knows?

Larry Thursby:

Unless they just pay their bill. I do my water bill, right? I’m sure the water companies send me all kinds of notifications.

Louisa Harbage-Edell:

You’re not my husband who digs in and sees what we did every single month, how our usage changes.

Larry Thursby:

Oh, if it goes up, I turn the water off. [panelists laugh] I know where the hot water valve is. I’ll just shut it off.

Louisa Harbage-Edell:

But I do think we talk a lot about that controllable issue and that sense of control that consumers get, which I think is what makes it more understandable and logical. Right? So maybe that feeds us into this idea of whether you call it telematics at point of sale or telematics on demand, but being able to actually have telematics data when you quote a policy or when you bind that policy. So Gina, I know you’ve done a lot of thinking and work around this. Do you maybe want to talk a little bit about where you see that? I don’t think it’s as far off as maybe we thought it was even six months ago.

Gina Minick:

Yeah, I like to think it’s around the corner. We’ve all invested a lot of time and effort because we believe that’s the case. Actually met with one insurance company a few weeks ago. I remember them saying that telematics at point of sale was the holy grail, and I expect that. No one’s really surprised to hear that. Who knew that there was a holy grail for the insurance industry? But I think we can all agree that this is it. It’s why companies have invested in telematics for years. It’s why we see now all of the top 10 large insurance companies have invested in programs because we believe in this and we believe in getting access to that data. And we have talked a lot about how do we make it even easier for consumers and agents. Well take the onus off of them to have that data at point of sale.

We took the poll. Louisa, how many people already have phones that are sharing their location data? If you’re using Google Maps, you’re sharing your location data. We already have that data. Again, we’re probably a biased group. I have 10 apps on my phone. I’m sharing data, but how can we take that data and harness it and collect it from different places? It shouldn’t matter where we get the data from. Robin mentioned earlier connected cars that long-term we are going to have connected card data. Take us some time to get there. But in the meantime, mobile data is becoming easier to access and it’s like the stars are starting to align. We already teed up. Consumers are more willing to share if they see value, they’re more open to sharing that data because they know they’re driving less. So if we can find a way, tap into that data and make it available for insurance companies, I don’t know who wouldn’t raise their hand to say, “I’ll take that data.”

Louisa Harbage-Edell:

Can we share up the results of that first poll?

Robin Harbage:

Pulling it up? Go ahead.

Louisa Harbage-Edell:

So over 50% of people have said yes that they have continuous location sharing on about almost 25% of you think you don’t. I’m really surprised if that’s truly the case, but maybe you’re a lot more sophisticated than me and you’ve actually turned off continuous location on things like Google Maps and all of the things that run in the background that usually share that. The 7% of you that are unsure, I guarantee you’re probably sharing location, but I mean this is the thing, right? So if less than 10% of policies are telematics, and the reality of course is you don’t take that data with you if you change insurance companies, over 50% of people, we know that something like over 80% of US consumers own a smartphone. Most of those are sharing data. I mean that’s a big difference in terms of being able to access that type of data. What were you going to say, Robin?

Robin Harbage:

I guess I was just going to emphasize that when you think about sharing data, I would agree with you that of the people, the non 54%, many of them are just wrong. And I know this for a fact because I’ve actually had the privilege of looking at data from a very large tech company, I think a global company that is almost a household name, and in that data anonymized and provided in a way that we could look at the possibility of scoring with telematics. I was shocked at how ubiquitous it was. And so to think that that data isn’t out there, yes, there’s permissibility who gets to use it, that’s a key issue to get around. So we protect people’s privacy, but the data’s there. And once you make that leap from, “Hey, I as an insurer have to collect my own data too,” there’s just data that is out there in the industry in the cloud that I can pull if the consumer allows me to, it checks all the boxes. It makes the ease of use for the consumer, it makes it ubiquitous so they can get it at point of sale and not some proxy or participation variable. It can become the major way in which I go ahead and rate my insurers. It’s huge in terms of, as Gina says, the whole degree up. It really will change the way we look at insurance as a means of delivering the product, the services, the safety – I really feels right around the corner.

Larry Thursby:

I agree. I think with that said too, there’s going to be a lot of work for us to do as insurance companies to be able to explain how it impacted. I can think back to when we implemented credit and as one of the folks that worked on the team doing that, I think one of the most common calls I had to answer was from an agent who said, well, I added or I pulled my credit for insurance and it changed my rate to make it not as low as it could be, or maybe higher than I thought it should be. And so how well positioned are we going to be able to, when it’s at point of sale, say to someone, why is my insurance cost more with you than it did with insurance company A, B, or C? Right? It’s a good conversation because we think we should be able to explain it based on your behaviors and how you can change those behaviors. But still, it’s a new conversation, right? Today, the independent agent doesn’t try to explain why their rate’s different. They just give it a business to another company, which maybe it’s good if you’re the telematics company and you’re avoiding someone who should have a higher rate, but at some point we’re going to have to make sure we’re able to explain that really well. So it’s easier and the consumer doesn’t need to do anything. But we still have some new things to do ourselves.

Teresa Scharn:

It’s funny, I was going to say that what Robin was explaining was that a couple of challenges are said at the beginning was customer adoption and customer interest and having data at point of sale certainly gives them tremendously, it makes it easier. So it helps simplify it and they can get even more of an immediate benefit potentially. So if you’re safe driver, it’s not the 10% participation anymore, it’s the 25% discount instead. And then I was going to say, it makes it easier for agents as well, but then of course Larry pointed out correctly as always, that some things can be easier. While there’s the benefit to them of having ideally on certain customers a more competitive rate. There’s always challenges associated with rolling out new solutions and always questions that you have to be prepared for or try to be prepared for as best as you can.

Larry Thursby:

I can tell you right now, my parents are nationwide customers to ask why they didn’t get the biggest supporting them to you, Teresa, to you. I’m not answering that call.

Robin Harbage:

One of the interesting things from the consumer standpoint that we haven’t addressed is portability of their data. Because today that data is locked up inside the insurer who they first start doing telematics with and doesn’t go anywhere from there. So if we start talking about this point of sale data that many insurers can tap into, now that data is portable for the insured to go from insurer to insurer and shop for the best rate. So as Larry said, if you don’t have the best rates from telematics, you may want that person to go to another insurer because they may be a bad driver or the next insurer may view that differently and they may have a better rate for them. So for the consumer standpoint, I think they will like the portability of that data if they have control over it.

Louisa Harbage-Edell:

In fact, we actually have a question in Q&A. I know we said we’d save them to the end, but it’s relevant here. So are we aware of any platforms or models where the insureds do own their data and could decide to share it with an insurance company? So I’m thinking of the Fitbit type model right now.

Teresa Scharn:

I think it’s largely with OEMs. So when insurance – GM, for example, if you are a part of OnStar or enable GM to with OnStar provides you those services, of course you’re sharing your data for those benefited services. They have customers then have the ability to leverage that data at point of sale with different insurance companies today. And there are other OEMs doing that as well. So the place that’s starting today at least is where customers own their data is really with vehicle manufacturers. And if you think about it, data at point of sale completely changes the game. And that one, it’s when segmentation and your ability to rate accurately becomes even more powerful. So it’s beneficial today because you’re doing it more on a renewal basis and we’re giving them a 10% participation discount upfront, but you’re able to more accurately segment upfront and attract the drivers that you know can write from a profitable perspective.

And then two, the other thing it changes as you mentioned, is the importance of retention of those customers. So if they can take that data and take it somewhere else today we provide people up to a 40% discount, they’re given an average of 23% discount. We know that it’s very difficult for them to go somewhere else and get that 23% discount tomorrow when they can move that data with them. They can. And so the experience and value added services, those things that we need to, that’s where we’re investing in today, is to create some of those experiences and go to continuous type of experiences to provide those value added services because we know that retention of that customer is going to be not just based on that rate, but also based on those experiences we’re enabling.

Robin Harbage:

I think one thing about the ownership of the data that’s interesting is when you look at the models that are out there for the large tech companies that collect data with a mobile phone and the mobile apps where that data is oftentimes used isn’t at the point of sale. It’s pushed upstream from that. It’s at the point of marketing. And one of the things we haven’t addressed here is what happens when the data is so ubiquitous that your marketing program is geared towards telematics and pushes it up to the point where now in the right situations, the best drivers are the ones to whom you market.

Louisa Harbage-Edell:

Yep. So Gina, I mean Arity does that today. Do you want to talk a little bit about how that experience works?

Gina Minick:

Yeah, I agree. I think this presents a great opportunity for low hanging fruit as we look of where we’re in the curve of telematics adoption with when we don’t have that data broadly available and broadly harnessed to leverage. When we do have that data, and we do know who the safe drivers are, it’s something we’ve invested quite a bit at Ity in because we believe strongly where we can find safe drivers and help insurance companies use their advertising spent to advertise to those safer drivers. We know those customers, first of all are going to have a higher lifetime value over three times the lifetime value of the average customer. So insurance companies want to target those safer drivers. I think Teresa already mentioned safer drivers tend to retain longer as well, and we know the safest drivers become even safer when they sign up for telematics. So this is such a compounding effect and really demonstrates the importance of if we have tools in the industry where we can target those safe drivers when they might be shopping for insurance and grab them now, there’s so much value in those customers in the long term. So those are the people that I think there’s a lot of value in targeting right now, right, with where we are in the trajectory of telematics.

Louisa Harbage-Edell:

So that takes us through all three of the major pieces we had in our original poll question. Are there any of the maybe other things that we didn’t have outside of the replacing traditional variables, telematics for targeted advertising, telematics at point of sale that any of you would touch on or highlight?

Robin Harbage:

There was a question that came up in the Q&A that it was kind of interesting, it’d be nice to address. It talks about today that individual OEMs and individual insurers are producing partnerships. And I think the questioner is correctly saying, well, my point earlier was it’ll have to be the many to be effective and ask how that actual fact that those things or those partnerships are occurring affects that. And I personally actually introduce one insurer and one OEM to each other early on in the process and said, you should be talking to try and encourage that process. But again, that insurer has less than 10% market share and that OEM has barely over 10% market share. So they can’t in and of themselves create a ubiquitous set of data that they can use for a large group of insureds or the public in general. And that is what I think is the limitation. So yes, I think that partnerships will occur, they’ll be very effective and very powerful for targeted sets of customers, but to expand beyond that much, you’re going to have to have larger conglomerated data sets in order to be effective.

Teresa Scharn:

We work with both today at Nationwide. So we have one-on-one partnerships with OEMs and we have targeted reasons for those partnerships. And we also get data through what we call these telematics data aggregators, which will help as Arity does. And other carriers we talked about do enable that many to many relationships. So we don’t have an intention of working directly with every single OEM separately or every single data source out there separately. We have an intention to doing that in specific cases that makes sense for us and for the consumer, but also we plan to leverage those telematics data aggregators and are already doing so today to make it more simple. So even though you see the OEMs having one-on-one relationships, they’re also sharing their data through telematics data aggregators to enable it to be available to the broader insurance group, which makes it much more portable across the insurance industry.

Larry Thursby:

And for what it’s worth as [inaudible] X or as Nationwide, it…I looked out at the market and thought, what is this company’s strategy or what is that company’s strategy? And you thought they were thinking coherently, at least at Nationwide when I was there. Teresa has changed this. To some degree, we just wanted data. So we would form a relationship with an OEM, we would with another third party who had aggregated data from other sources to learn, right? So if you looked in at Nationwide and said, I wonder what Teresa’s up to, I wonder what she’s trying to accomplish. Undoubtedly there are papers that say, here’s what this relationship means with this OEM, but to some degree a big part of it’s just experimentation and getting the data and getting used to it because it looks a little different from every source. And so again, just so folks here don’t feel like, well, gosh, I’ve got a discount. I’m working with one cell provider, one smartphone app, am I doing it right? Probably the answer is you could do more, right? That’s always an easy answer. You could do more, but other folks to a large degree are trying to figure it out, but with a driving purpose and at least in the case of Nationwide, just to get more and more data to understand and be prepared for that future that I think we all see where data is everywhere, but by the time you get there, are you ready for it?

Robin Harbage:

I think another issue that we’ve obliquely talked about, and that is cost. I know one of my former employers, one of the keys to being successful was always driving cost down. And they were very successful at that in many of their business practices and no less so in telematics, always looking for a way to lower the cost of driving a telematics program. And that cost efficiency led to lower prices led to growth and profitability. As you think about the data becoming ubiquitous, coming from many sources through something besides a device that has to be sent out or an app that needs to be maintained and downloaded by the insurer, those cost efficiencies can be passed on to the insured as we move forward in this. I think that’s probably more critical than people realize because it’ll be a competitive game where people can lower 1, 2, 3, 4% of their expense ratio because they don’t have to go out and individually try and pull all of the data together or maintain devices or apps.

Louisa Harbage-Edell:

Excellent. Well, I see a lot of questions coming through the Q&A panel, so maybe we’ll put a pin in this and wrap up and pivot to that. We’ve talked about a lot here today already. We’ve chatted about a little bit of the history of telematics and you should based auto insurance and how it’s evolved. We’ve talked about the ways that kind of data can help us understand changes in our book of business and our consumer’s behavior. We’ve talked about some ideas on how to improve current program growth from distribution to consumer adoption. We’ve talked about some exciting insight into where we think telematics is going and may take us in the near future, most notably with telematics used for acquisition at time of quote. So with that, let’s turn ourselves to the Q&A panel here, and there are quite a few, so lemme just scroll up. We do have one stating back to our COVID conversation about how the regulatory environment may have changed with regards to telematics as a result of the pandemic.

Teresa Scharn:

I have to think about it as a result of the pandemic in general. I think the regulatory environment has been extremely favorable for telematics in that it is much more understandable to departments of insurance and to consumers that you would be rating based on how people drive. There’s always individual state cases that are unique and they’re always the same states that are very different. But in general it has, even with our pay per mile program, it has been being one of the first in many states, if not the first, in many states, it’s been fairly smooth and fairly simple for changing something that much in insurance. And I think it’s because it makes sense, right? It makes sense that how much you drive and how you drive are part of your rate. And they seem to, it is been actually pretty, I’d say pretty calm, pretty simple.

Larry Thursby:

Hey, Teresa, from a rate filing perspective, so a little bit separate than telematics, just your common run-of-the-mill, everyday base rates. Are regulators challenging you about using your past three year trends to set a rate for the future saying, Hey, we know people are driving less, so we’re not going to allow you to file this plus 3% indication, we need you to select a more recent trend. And are you able to use telematics data to bounce back or is it just really all quiet in terms of you’re able to push through the rates that you need without any obstacles in terms of folks saying there’s less driving?

Teresa Scharn:

No, that’s a fair point. There are definitely obstacles in that space around getting, if you have trends or overall trends that seem to be positive in states or needing to take slight rate increases or even desires for us to take potential rate decreases due to reduced driving, there is definitely a rise in questions coming from departments of insurance around, prove to me that you, from a trend perspective and how is your telematics data or how is your data showing that you should be looking at the shorter term versus the longer term? And has it come back to normal? Have the trends come back to normal? Have they not come back to normal? I’m not articulating this well, but in general it’s we’re getting more questions about looking at more recent trends instead of using the in general three-year trend and we’re leveraging our telematics data to work to tell that story.

Larry Thursby:

Good. That’s another, right.

Louisa Harbage-Edell:

So Gina, a follow up on your comments about the consumer engagement and their willingness to share data. Was any of that broken out by age group?

Gina Minick:

Yeah, I don’t know if that study particularly by age, but we do see that younger consumers tend to be more open or more willing to share the data. Again, it’s because they see value in sharing that data. So I’ve definitely seen trends skewed towards the younger generations being more willing to share.

Louisa Harbage-Edell:

But to your point, that specific piece of research that we did was not, but we’re getting ready to run it again. And it’s an interesting point that we may want to think about adding that variable.

Teresa Scharn:

When we look at Nationwide at our distribution of non-telematics business and our distribution by age of telematics difference, the two look more similar than not. So there is a little bit of skew towards younger drivers, but it’s not to the extent that you would expect. It’s just maybe 2, 3, 4 points difference at the higher levels versus lower levels. But it resonates with drivers across all ages, all because there are safe drivers across all ages. There are people who have lower mileage across all ages and you’re able to have a message that resonates with them across all ages.

Louisa Harbage-Edell:

I think to some degree, this gets back to the point you all were making about education. I’m just making sure that our insurers understand the benefits and what they’re opting in for, right?

Larry Thursby:

And I think our agents too, right? If an agent believes that a younger person is more open to telematics, they might be more apt to offer it to that younger person. So it’s making sure that we have the data so that we can help the agents to effectively sell too.

Teresa Scharn:

That’s why I have the data.

Louisa Harbage-Edell:

Absolutely. So we have a question here about whether there are any good tools for combining data from all of these disparate sources.

Gina Minick:

I can give a less technical, try to be a non-technical in my answer because at Arity we are ingesting data from many different sources. I like to think we’re sensor-agnostic and OEM data from the vehicle is going to look different than data from a mobile phone, for example. And it’s not just because the GPS trail might look a little different, maybe there’s slight nuances in the calculation, which we can actually reconcile and streamline that piece of it. But there might be different data like distracted driving, we get from the phone and we’re going to get adas from a vehicle, for example. So when we think of ingesting that data, we put it in a common format, but we’re still going to have layers that are going to be unique to if it’s more driver-based or vehicle-based.

Robin Harbage:

So one of the things I think people need to understand, they think about the disparate sources, understand that a Google phone, an Android phone and an Apple are calibrated differently. So when you think about disparate sources of data, it’s not just a question of an OBD or an OEM or a phone, it’s even within phone types and phone styles. I would even, although I’ve not looked at it in detail, believe that an Apple [iPhone] 12 versus an i8 probably have different calibrations. So this is a task that’s really technical and at the base level that people have to understand to get accurate data that they can calibrate across different devices to be fair to the drivers.

Larry Thursby:

And I think too, it’s a great point. Regulators have generally favored telematics programs, as Teresa was saying, but for insurance companies to be ready to be able to address those types of questions to a regulator, right? If I’m the department of insurance, I don’t want to hear that the i12 and the i8 might give me two different answers for the same driver. And so as insurance carriers who have telematics programs being prepared to proactively address those types of good questions, that’s a great question that we should be ready to answer.

Teresa Scharn:

I think this is a key area where don’t be afraid to partner either partner with a mobile provider or partner with somebody else, but what are you trying to differentiate on? And for us that’s being able to ingest data from multiple sources quickly and what is the best way to go do that? It might not be tech, it may be a partner external.

Louisa Harbage-Edell:

So partnering with OEMs, we have a question here about whether OEMs may become insurance brokers in their own right. I know we’ve seen at least one major announcement in the last couple of weeks for that effect.

Teresa Scharn:

I mean I think the answer OEMs have been insurance brokers in the past, right? This isn’t necessarily new and definitely an area of potential for the future. I think as vehicle technology changes as maybe it becomes much more autonomous, insurance might become part of the vehicle. And that’s the world that we don’t know how that’s going to shift over time, but that’s the world where it’s certainly becomes a much more viable solution, certainly long-term, but there’s a lot of challenges with that short term as well. The challenge is does somebody have a Honda and a Ford and a Toyota all in their household? They don’t have, so can you go to Ford and go get insurance for all three of those vehicles? And then does a consumer want that experience to vary between all three of those vehicles? And so in today’s market, there’s challenges associated with that that we’re all working to consider and determine how we get past some of those and understand where the future’s going. But they’re already becoming insurance brokers and actually already were before.

Louisa Harbage-Edell:

I think, Robin, you’ve made an interesting point in the past, which Teresa, to your point about the long-term and the autonomous vehicles is where does that blurred line between personal liability and product liability occur?

Robin Harbage:

This is an issue that I’ve pushed for some time and approached OEMs with whom I worked as well as insurers to think about. And that is the product liability of the car makes an error and the car causes a crash versus the person makes an error or two very different issues and who’s responsible for that? It is a little farther out in the future, but I think we’re going to have to think about a day in which the insurer and the OEM have some kind of agreement between them. It’s almost like when you sell a clash coverage, two insurers, excuse me. An insurer has two cars that cly with each other. Both of them are their insurers. They assign separate adjusters to each of those claims, each of those drivers. The reason being they want each one to have this own advocate and then they settle a lot of the issues behind the scenes without having to the insured through the issues. There’s a long way of explaining the OM and the better have a way of negotiating that liability and that personal liability issue to the of.

Louisa Harbage-Edell:

We also have a question here about best practices for managing data security. It’s obviously a major point of concern for consumers. We want to make sure we’re doing the best job possible. So any thoughts on best practices or things you’re doing now? I mean, Teresa, I’m sure you’re thinking daily about some of these aspects.

Teresa Scharn:

Yeah, I think one of the things, the experiences for insurance companies, we’ve always had to protect data and manage data. We’ve always had data that we’ve always needed to have very strong protections around and this isn’t any different. So it’s continuing to build those strong data protection capabilities, managing that very closely and we take that very seriously and have for years. And then two, it’s just being very clear to consumers what data you’re getting in our experience and what the opportunity we have with the experience is to show them the value of the data and what that data actually means. And anyway, I would say we’re just really, really focused on continuing to protect them, protect the data as we always have. This is not a new space for insurance companies.

Louisa Harbage-Edell:

Absolutely. We also have a question here about the role of telematics in sharing economy or shared mobility. Gina, it’s obviously a space that Arity plays in as well. Do you want to speak a little bit to that?

Gina Minick:

Sure, yeah. It’s interesting when we think about the insights that we can derive based on how people move and that risk associated with it, it’s so relevant for personal lines, but those same analytics really hold true when we look at the movements in the car and rideshare space with just a little bit of a twist. They’re slightly different problems they’re trying to solve, which is really often the problems that I see with my counterparts who are managing that area are around how do we continue to bring in a safe driving pool? How do we grow the supply because the demand continues to be there. COVID has actually presented an interesting opportunity with people using these car share services, more delivery drivers. We see even with that dip I mentioned earlier with personal was really around personal lines. We see less personal lines. We did see less in that car and ride share space, but not as less as personalized.

We still have quite a few people out there in that industry and they’re trying to solve some really hard problems to really manage that pool of risk, bring in safer drivers and help understand what their potential loss costs will be. So I think it definitely plays a huge part. And also if you think about the experience and the driving experience in that space, there’s a lot of overlap between how I drive as my own car versus how I drive if I’m an Uber driver, might not be that much different. So there’s a lot of learnings that we can kind cross-pollinate there.

Robin Harbage:

So as a consumer, he use ride sharing a lot before Covid. I personally would love the transparency of before I get into an Uber or a Lyft car to see a ride share safety score on that driver because I’ve gotten into a couple of cars for a rideshare where after the trip I would’ve never gotten back in that car if I’d known what kind of score that person was going to generate during that trip.

Gina Minick:

Yeah, I’m dying for that to happen too because you can, right? I’ve been in a foreign country before and I’ve paid extra for someone who speaks English because I didn’t speak the language. I would pay extra in a second for someone to drive me safer. Again, I’m biased, but we have the data, we have data to do that. So I look forward to the day when we can use our data in that way.

Louisa Harbage-Edell:

Excellent. Well, I know we have a few questions we haven’t been able to get to, but I’m conscious we are also at the top of our time here. So I’d like to thank all of our attendees as well as our fantastic panelists for joining me today. I certainly believe telematics already has and will continue to revolutionize the insurance industry. And after today’s session, I hope you all agree. Thank you so much.

News & insights

blog post

What is Arity Insights? 3 themes from the mobility data ecosystem

Three key themes from our Arity Insights event in Scottsdale, AZ.
blog post

How telematics is revolutionizing auto insurance marketing strategie…

Arity’s Jen Gold shares five ways telematics is changing the way auto insurance carriers approach growth.
news

Car Touchscreens: Oversized, Over-Complex, or Useful Driver Systems?

"Are modern car touchscreens getting so big, with so many sub-menus, that they are a safety hazard? Should insurers fact…
Insurance Edge
news

Can telematics drive fairness in auto insurance by reducing bias?

"Traditionally, insurers rely on demographic markers and even credit scores to assess risk - variables that can unfairly…
Insurance Business