Do drivers who wear turtlenecks get in more accidents than drivers who don't? What if the theory proved true? Although it would be nice to find out whether this level of minutia affects accurate pricing of risk, when it comes to ratings, this question is not one insurers are asking-yet. But insurers are finding unique ways to price risk-from aggressive braking patterns to career choice. Progressive is one such insurer taking steps toward accuracy, and it's using technology to do so. Most recently, Mayfield Village, Ohio-based Progressive Direct Insurance Co. launched an odometer-reporting, Web-based pilot program in Iowa and Virginia. And, in the summer of 2004, the carrier began piloting a usage-based program in Minnesota called TripSense.Using a device called a TripSensor, which plugs into the ODBII port in newer vehicles, Progressive is collecting driver data in the hopes of creating a win-win for its underwriters and its policyholders.


The odometer-based program enables Progressive customers who volunteer for the program to periodically submit odometer readings-via a password-protected Web site-on the insured's vehicle when they sign up for the program and at each renewal of the policy or every six months.

Progressive then annualizes the reading and uses that mileage to dictate the discount, which can result in up to a 15% reduction in premium per vehicle for a six-month policy.

Some insurers ask their customers to estimate the number of miles they drive in a certain amount of time, but Progressive's Iowa product manager, Ian Forrester, thinks the new odometer-based program will eliminate guessing.

"Typically our competition will ask, 'how many miles do you drive in a year?' or 'how many miles do you drive to work?' Customers don't necessarily know how many miles they drive in a year," Forrester points out. "And if their situation changes-say they move or change jobs-that mileage might change a lot."

Before developing the odometer-based program, Progressive collected information from the Ohio Department of Motor Vehicles, including odometer readings it received from license transfers, according to Forrester.

The carrier matched that information to pre-existing customers to get some proxy for the theory that loss performance is connected to miles driven. And by piloting this program, Progressive hopes to sort out remaining issues, such as how accurate the data really is and how reliably customers will report their data.

Though Progressive is not yet ready to release the exact number of participants, Forrester says there's been significant interest in the program, which officially began last December.

Some of that interest may be related to the 5% participation discount on a vehicle's total premium, which customers receive when they sign up and continue to receive if they continue to report their odometer readings.

Why the discount? Forrester says Progressive wants to collect as much data as possible and hopes the discount will interest all customers no matter how many miles they drive. Even the customer who records the highest mileage in Iowa or Virginia still receives the 5% discount.

"I wouldn't be surprised if we stick with that as our long-term price structure," says Forrester. "The problem you have with usage-based insurance is until you have someone driving and you're collecting that information you don't actually know what their performance is."


It's been almost two years since Progressive began piloting the TripSense program. And since then, the ODBII port has become ubiquitous in the market, chiefly because the U.S. Environmental Protection Agency mandated its installation in new vehicles for emissions testing and monitoring purposes, notes Dave Huber, Progressive's TripSense project manager.

The ODBII port communicates with the majority of electrical components in a car. Therefore, Progressive realized the port could be used to capture data relevant to improve rating accuracy.

"Progressive and the rest of the industry have access to certain kinds of (driver) data. From an application perspective, you can obtain age and marital status, where they live, what kind of car they drive-things that can help classify different groups of individuals," says Huber.

"And we also have access to driving history from the state-accidents, speeding tickets, claims made. Any piece of information you can obtain to figure out whether this person is more or less likely to have an accident allows you to more accurately price [a policy], and from a competitive perspective, this might allow you to acquire more business than the competition."

Huber believes if other information becomes available through TripSense-such as whether drivers wear their seatbelts all the time, how fast they drive, how much they drive, how aggressively they press on the brakes-rating becomes even more accurate and policies can be priced accordingly.

This benefits Progressive and its customers, he says, using an example of competing for a customer: He and his competitor know the same things about a particular customer, and so their rate is the same-$600.

"But if I can figure out something else about the customer (such as the information provided by TripSense) that leads me to believe that person is less likely to have an accident, then I'm comfortable offering a $450 rate instead."

However, what about higher-risk drivers, whose data may show are more likely to have an accident? "Rather than the $600 price, I may raise my price to $700-because that's how I've assessed the risk," says Huber. "So that particular customer may choose to buy from a competitor at the $600 rate."

But if the data about the higher-risk customer is correct and that person has an accident, the competitor then may have to raise their rate, not to $700 but to $800, to make up for the loss, he says.

"And if that's the case, I am still sitting at $700-my initial offer. So, if the market is really efficient, that customer is going to find my $700 offer very attractive and come back to me anyway."

Those types of "vitals" are what the TripSensor is measuring, even outside Minnesota. "We continue to pilot the TripSense product and discount in Minnesota and have about 4,000 or so Progressive Direct customers who have elected to participate," says Huber.

"We've also sent e-mails to existing customers across the country and said to our customers 'we're doing research, and we'd love it if you'd help us with that research,'" he says.

The TripSense discount is the sum of the usage discount (which begins at 15% and is reduced by 0.00085% for every low-risk mile, 0.00125% for every medium-risk mile, and 0.00700% for every high-risk mile driven per year); the safety adjustment (which is based on the percentage of non-idle driving time spent above 75 mph); and a 5% upload bonus.

Those customers who volunteer are directed to a Web site to register. The receive a TripSensor, which they plug into their car, and and they can upload the date onto a Progressive Web site by plugging the TripSensor into their home computer. Upon receipt of the data, Progressive sends $50.

"So we're buying the data, in other words," says Huber. It's not tied to the insurance product. It's not tied to the insurance premium. It's not even tied to the customer relationship."

So far, about 20,000 Progressive customers across the country have volunteered to provide information.

"We're essentially trying to collect an 'earned car year' of data," Huber says. That might be obtained via 20,000 devices in cars for one year or 10,000 devices in cars for two years."

Huber hopes that by year-end, the data collection from those 20,000 devices will be enough to prove theories "wrong" or "right" and figure out a way to implement the "right" ones.

Register or login for access to this item and much more

All Digital Insurance content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access