Technology has infiltrated every aspect of the insurance industry, but nowhere is it so important than in the area of usage-based insurance. Insurers have been looking at telematics devices for almost 20 years, but one carrier—Progressive Insurance—has staked its claim as the clear market leader.

Through 2011, the biggest challenge Progressive faced with its telematics program was finding technology that was inexpensive and easy enough for customers to use to make it a viable product, according to Dave Pratt, general manager of usage-based insurance for Progressive.

The insurer first piloted a UBI program in Houston back in the mid-1990s. Then as now the idea was pretty simple: If you could measure how people actually drive it would be a fair way to set insurance premiums instead of using traditional data such as age and marital status.

Progressive conducted a pilot program with professionally-installed devices that would measure how customers drove. The data was good, according to Pratt, but the device was expensive and involved too much work for the customers.

Progressive moved on to an OBD plug-in device and Pratt explains customers were supposed to plug in the device, leave it in for six months, and then find the plug the carrier sent with the device six months earlier to plug it into a computer. Again, too much was being asked of the customers.

Finally, Progressive developed an OBD plug-in with cellphone technology built in so it was easy for the customer to transfer the data, according to Pratt.

“Plug it in and forget about it,” he says. “After every trip, the device calls home and sends the data about that trip. The testing on that model showed it worked well. It was easy for people to do. The next challenge was to explain this new technology to people. We spent a big effort in marketing beginning in 2011 to advertise Snapshot and that built up an awareness of the product and as that built up we’ve seen more people use it.”

Today, Snapshot is the market leader and while the carrier has been able to attract good drivers to the program, some believe the carrier has not yet capitalized on the secondary aspect of what telematics has to offer—improving the driving skills of its customers.

 “Progressive is are trying to find the best driving risks, but they are not set up to provide improvements in driving behavior,” says Karlyn Carnahan, research director for Celent. “Some carriers that have been going into telematics have been focused on an opposite proposition: how to improve driving behavior, reduce accidents, reduce claims, and reduce the amount of premiums paid.”

One of the things that excites Pratt about the business is it is new and evolving. A couple of years ago Progressive added an audio feedback feature so after a hard brake the customer hears a beep.

“Getting that immediate feedback trains those drivers to be aware of situations when they have to hit the brake hard,” he says. “We are confident customers have had fewer accidents because of that feedback. I see that as driver feedback version 1.0. If we can help people drive safer that could be a powerful change for the insurance business.”

A great deal of the success Progressive has forged has come from its marketing efforts and Pratt doesn’t believe that will slow down anytime soon.

“We need to continue our marketing effort so consumers recognize Progressive as the company that has Snapshot,” he says. “We need to take advantage of the lead that we have and the mind for insight that allows us to stay ahead of the competition. If we know which drivers are the safest we can give them the best deal with Progressive.”

Slow Adoption Thus Far

A year ago there was much excitement over Progressive’s offer to sign competitors to licensing deals as long as the insurers held off on implementing a UBI program until June of 2015. Richard Welch, vice president of corporate planning for Concord Group and a close observer of the telematics business, expressed surprise that more insurers didn’t take advantage of the Progressive offer because Progressive wasn’t charging a big price.

Patience has paid off for some, though, as Liberty Mutual won a patent case against Progressive in late March that struck down three of their seven patents, although Progressive announced it was appealing the decision.

“Insurers weren’t necessarily sitting around waiting for a court decision. But I thought there would have been more movement,” said Welch. “On the other side, Progressive finally succeeded in proving the business model. About one-third of their new customers are UBI customers. That’s very significant. The doubters felt there wasn’t a big enough portion of the insured public that would want to do this. I think that points otherwise.”

Welch believes Progressive is sitting in the catbird seat.

“By the time the rest of industry starts to catch up, Progressive will have a huge lead so it will be interesting to see how that plays out and what it means,” he says.

Carriers didn’t join with the Progressive license because telematics is expensive to begin with, points out Carnahan, and if insurers had to pay a license fee and go through a waiting period, that just added more barriers.

Chad Hersh, managing director of the insurance practice for Novarica, believes there is a smaller push by consumers for telematics, in part because in the U.S. there are much weaker privacy regulations compared to Europe. In the UK, where telematics is extremely popular, people feel comfortable giving insurance companies their information because it is well protected.

“There are other carriers starting to introduce telematics products,” says Hersh. “Some are working around the Progressive patents, some have sued Progressive and won. There is a fair amount of opportunity to really go after that market. However, they have to be mindful of the fact Progressive got the first wave—the best risks. You have to be careful.”

Part of the challenge for insurers is UBI relies heavily on technology, so it’s something carriers have to prepare for and some insurers aren’t there yet.

“The billing system has to be prepared for it; your policy system has to be ready for the products to be developed; your underlying data warehouse or analytics have to be mature; you have to understand the patterns and what discounts will apply,” says Hersh. “There just aren’t enough insurers there yet.”

UBI is optional for Progressive customers and the insurer tries to price the both the UBI product and the traditional personal auto policies so profit margins are the same for the people who sign up for Snapshot vs. the people that don’t.

“We don’t want the people who do UBI to subsidize the people that don’t,” says Pratt. “Progressive doesn’t want a lot of adverse selection. I hope if we still have the best UBI program we attract more of the good drivers to Progressive. In my world it creates adverse selection against the competition, but no such thing within our own book of business.”

Hersh believes there was a natural market for UBI. The people that are the best risks and the early adopters were going to be the ones to sign up initially.

“It’s why you come up with a product like that—something that rewards being a good risk and allows you to pull in a pool of great risks,” he says. “After that initial rush it’s not necessarily clear how much you want that rush to continue. Once you have the folks that self-identify themselves as a good risk and you go through that list it slows down.”

Trouble for the Midtier

Telematics is a major issue for midtier carriers, but those Carnahan has spoken with have smaller R&D budgets available.

“How does a $250 million insurance carrier compete against a $5 billion carrier when it comes to investment?” she asks.

Devices such as what Progressive uses cost money, explains Carnahan, transmitting the data costs money, storing the data costs money, analyzing the data costs money, sending back the device costs money, and putting together a new policy costs money.

“There is a huge array of costs associated with capturing and leveraging telematics data,” says Carnahan. “A midtier says this is a huge investment and when they look around they see a low take-up rate. Less than one percent of U.S. drivers are using UBI and 95 percent of those are using Progressive.”

Working with a mutual carrier, Welch explains that Concord is interested in UBI, but is not in the market yet.

“As recent events unfold and we look at what the next year looks like I think there is pressure on us to have something in the market and be there for the long term,” he says. “If you would make 100 of these calls to mid-level companies they would tell you they need to be there at some point but it is a question of when. There are competing priorities. For us it’s not a lack of interest; the impediment for us is when you look at our list of top five things we have to do is that number one or is it number five. If it is one, we are ready to jump all over it; if it’s five we have some other things to take care of first. We have to reassess at this point.”

Carnahan explains that if an insurer can find the answer to that question they might be able to generate a return on their investment to justify the expense, but she believes many consumers don’t understand what UBI is.

“Drivers have heard about it, but they don’t know how to get a hold of it or want to switch from traditional policies,” she says. “Some drivers have privacy concerns. Are they giving too much information? The carrier side is trying to get it into the queue and deal with organizational challenges and some of the legal issues. For a midtier there are a lot of operational challenges.”

There are other big players that are moving from pilot mode to a real program and they are ready to connect to their customers, but Welch doesn’t believe the industry is getting traction from players beyond the top four or five insurers.

“Others are either still in pilot experimentation mode or some may have commercial programs, but they are in few states and to me your program is not real until you are ready to advertise and push for sales,” says Welch.

That means the midtier is going to have to catch up, but Welch doesn’t believe that’s an impossible mission.

“You can start a program and get it in the market quickly, but this is one of those Big Data applications with lots of analytics behind it,” he says. “In the short run, I think you can run a successful program without being too far ahead as long as you have a measurement system to know who the best and worst drivers are so you can set a discount schedule. I still subscribe to the duel set of tipping points. One is for insurers that feel they need to be in the game. I think that’s going to take place in the next year. The second thing is there is some fear as to how big of an advantage Progressive is gaining through what they are already doing.”

Big Data

Progressive recently announced they have 10 billion miles worth of driving data, which the insurer needs to store and manage. With so much data, Progressive had to upgrade its technology and is in the process of installing a state of the art system that distributes the data across multiple server clusters to help manage it.

“Moving to the new technology allows our analysts to test new ideas faster,” says Pratt. “If an analyst has an idea for a pattern to predict safe driving, we can crunch the data and evaluate it more quickly with the new tools.”

The data Progressive has been collecting is pretty simple, according to Pratt. For every second of a driver’s trip, Snapshot collects the vehicle’s speed and calculates how many miles were driven and where there was a hard braking event or hard acceleration.

“With that simple data set, it wasn’t all that hard to figure out what mattered and what didn’t,” he says. “We are testing some location data. If you look at where you are driving, what type of road, the weather conditions, the time you were driving. It might be useful to determine if that was a risky condition. We’re just doing research on that this year. We’ll come up with theories and use the data to see if they are correct or not.”

In order to maintain its market share, Pratt believes the carrier has to leverage the Big Data advantage Progressive already has achieved to keep ahead of the competition in terms of the best models and finding the safest drivers. Progressive also has to pay attention to emerging technology that might be useful.

What are analysts learning from the data?

Hersh believes there is enough data to start making good judgment calls on driving patterns. One of the challenges, though, is unless you band together with an ISO-type group for smaller carriers, it could take a long time to gather enough data to make the right decisions.

“You need data scientists to make good decisions,” he says. “The bigger you are the quicker you can get to the critical mass of data. I suspect that outside the UK, Progressive is the only one that has that data today. With the initial filings you are making educated guesses. It may turn out that the people that really have an extremely high rating may be the people that have gone to driver’s training school and enjoy pushing the limits of their car and know exactly what those limits are. It’s all educated guesses until carriers get some experience and there are some carriers that are quite content to let Progressive see if it does work.”

Data is an enormous issue for anyone contemplating the use of telematics. There are many questions to be answered such as: How do you collect useful information on driving behavior? How do you collect this extraordinarily granular data and turn it into something to use and more importantly something you can analyze?

“You can validate loss trends over a period of time, but carriers should be able to judge severity, develop response protocols, reduce fraud, and reduce rate evasion— how many miles a particular motorist drives,” says Carnahan. “Telematics should provide a faster reporting of claims and therefore the faster you report them the less they cost. The promise of telematics data is enormous, however the reality is they can’t fully take advantage of it.”

Data and business intelligence are among the top areas carriers are trying to address, according to Carnahan. Big data is not being used widely, but carriers see enormous potential in it.

“Most carriers are not able to take advantage of the data provided by telematics,” she says. “Driving scores is one example to use as part of loss analysis and rate making. Carriers are looking at putting together more robust tools and capabilities so they can better utilize the data, which leads to the fact there is not a plethora of data scientists in the insurance industry. Carriers need to think how they can attract these people or use third parties to aggregate data scientists and utilize them for specific projects.”

Data Analysts

The biggest area where their the job of a Progressive analyst has changed is they have to spend more time thinking about the quality of the data and making sure the conclusions are correct.

“In the traditional personal auto product, if you want to base the rate on age and marital status, that’s pretty simple,” he says. “Let’s say if we want to find how many times you have to hit the brakes and you want to build a statistical model for that, you have to figure out if you are correctly identifying hard brakes.”

Early on, some Progressive customers would call in and dispute whether they had a hard braking event. One area involved snowy conditions where cars lose traction and the wheels spin.

“If all you do is a simple calculation without any quality screening you would say it was a hard brake, but it wasn’t,” says Pratt. “We need to build tools to identify those kinds of problems and correct them.  With the statistical modeling we did in the past you had to make sure the data was correct, but it was much simpler than collecting second by second data and analyzing it.

Miles driven is of great importance to insurers as well as the time of day you drive. But on the specific driving behavior Welch feels there are three things that deserve the most attention:

1. How often do you brake hard? Progressive has published their specific parameters for that and other companies are aligned around that standard.

2. There is the reverse of that data: rapid acceleration.

3. There are carriers that measure super-speeding incidents. Companies use GPS to measure how often you speed more than a designated number of miles per hour above the posted speed limit. It requires detailed mapping software behind the GPS system so they can know at any point what the posted speed limit is and what the actual driving speed is. They are not interested when you are going 70 in a 60, but 90 in a 60 or 80 in a 50. Those are big risk indicators.

    

   Adverse Selection

   Adverse selection should be a major issue for insurers, particularly if UBI grows and insurers feel the need to jump in without the right analytics capabilities.

   “UBI is not the type of product you want to do that with because that’s how you end up with the worst possible risk,” says Hersh. “If someone is a good risk you are going to show them the most favorable price you’ve got; if they are a lousy risk you’ll show them the worst possible price. That is a way to manage adverse selection. If carriers aren’t careful they will end with a risk pool that is not ideal. Without analytics capabilities they might not even know it.”

   The good drivers are going to go for telematics and get lower prices, which means poor drivers will stay on classic policies and the prices for those policies will rise, according to Carnahan. If carriers are not delivering the telematics experience they are going to end up with a growing core of poor drivers.

   “There is an assumption that drivers want classic policies and the ones picking telematics are new drivers or those that want to prove they are good. Or they are infrequent drivers who want to prove they are a good risk,” she says. “In this hypothesis you assume customers don’t want them to spy on them and capturing the data is expensive. One of those hypotheses may be more right than the other, but it is still too soon and we don’t really know yet.”

   Making Inroads

   There are a number of carriers making inroads with telematics and it tends to be the larger carriers, points out Carnahan. Smaller carriers like PURE and DTRIC in Hawaii have UBI programs.

   “The question is how to differentiate what you are doing,” she says. “There is more in the commercial world with fleet management. It gives you things like fleet tracking, roadside assistance and fuel management programs.”

   PURE has a value added service that includes a distracted driving solution to keeps drivers from texting or searching the web while they drive. American Family has a fleet program. Hartford and Zurich each have programs to help drivers.

   Gamification also has a place to play in the telematics landscape, adds Carnahan. The Evogi Group has a virtual world to set driving improvement goals and based on that behavior you achieve rewards. You can challenge other drivers to complete missions and compete for status.

   “The idea of using game-like technique to change behavior is a big aspect of telematics,” says Carnahan. “It gives you the chance to go on a fun-type application to look at your family’s driving or your teenager’s driving and monitor and measure it in comparison to others.”

   Using telematics in the commercial lines might help drive telematics in the U.S. more than personal lines, according to Hersh. Many trucking insurers are using telematics because the privacy concerns aren’t there.

   Fleet operators can look at driver behavior, save fuel, improve safety, and track mileage in ways that benefit their company and the driver doesn’t care about privacy because it’s not his own car.

   “The more carriers that get into the space, the more the consumers get educated and want that type of product, the more other carriers are forced to do it,” says Hersh. “Consumers willing to buy a policy from Progressive and are willing to do telematics and have driver behavior factored into their pricing may be tapped out. There may be a secondary market that does all that and want an agent.”

   In the next year, Welch believes UBI will take off at the insurer level as carriers enter the field and start to advertise that they have the product. Consumer adoption is going to move slowly at first.

   “There is potential to take 20 to 25 percent of the market over a five to seven year period,” says Welch. “Going beyond that is going to depend on one thing none of us knows much about: what is the behavioral impact this will have. All the modeling will stay on the idea that you can rank drivers as good or bad and that this program has a certain cost to it. There are economic trip points that you can attract someone to the program, but if you were to say once they are in a UBI program and they are being coached by the software how much better of a driver have they become? If that is a big number, the 20 percent could grow pretty big. If it’s a small number than it could eventually reach a third of the market but not much more than that.”