Car insurance even predates Henry Ford, yet a truly disruptive auto insurance provider hasn’t emerged in more than a century. Metromile CEO Dan Preston tells Karen Webster how his metered-model company is looking to disrupt the $288B auto insurance industry with smart tech, a new business model and a data-driven strategy.
It has been more than 100 years since the first car was insured in 1897 by the Travelers Insurance Company. Since then, there has been no-fault insurance, safe driver insurance and student driver insurance – but outside of that, no essential changes have been made in a $288 billion dollar business. Until now.
Based in Silicon Valley, Metromile is bringing a new business model to car insurance based on a per-mile metric. If you drive less than 12,000 miles a year and qualify for standard underwriting rules, Metromile will calculate your premium based on miles driven, not preset premiums.
As Metromile CEO Dan Preston told Karen Webster, the prevailing wisdom that the more you drive, the more likely you are to get into a car accident is still largely true. However, new technology is able to track the true number of miles driven by a car, and can even track how a driver handles a car.
With that new business model and technology, Metromile launched in 2016 and is currently available in eight states, which include California, New York, Washington, Pennsylvania and Arizona.
“I think it’s really exciting, how you can build an insurance company from the ground up, with sensors now being applied as opposed to everything being manual,” said Preston. “You start to think about how a company could service their customers differently. So I think we’ve become much more of a digital insurance company [that is] building an entirely different kind of way of doing business, which evolved out of that initial idea.”
Metromile is a full-fledged insurance company. It started business in 2016 as a broker and then bought an insurance company that effectively became Metromile. It has its own policies, sales staff and underwriters. With the unique per-mile model, it will not only focus on different KPIs, and pick out the safe drivers and safe roads, but it will also select the safer time of day when there are fewer accidents. All of these are factored into the price, and drivers get discounts for those conditions as well as their driving record.
The base rate, however, is not calculated based on how many miles the customer drives. There is a basic premium to cover collision and acts of God, which represent about 30 percent of a traditional insurance policy premium. The rest includes a number of other factors, like the number of accidents the drivers had, the type of card and the specific Metro area. So the basic underwriting is standard – but the biggest challenge from Metromile is not discerning who is driving and how they drive, but ensuring that the company and the per-mile business model as intuitive as possible.
“We need to balance that as one of the things that’s as important to us as we introduce new products. How do you make the product really intuitive?” Preston asked. “How do you tell that you are the most accurate from a risk perspective? Or how do you deliver a great product experience? These are the kinds of questions that we grapple with internally all the time.”
Metromile is at the point where it can start to use all the data gathered by the sensors in the car to draw conclusions about its drivers. It recently did a study that tested a free insurance program that allowed users to test-drive Metromile. It found that once a test consumer switched over to Metromile, they drove 10 percent less on average. The potential positive impact in a number of different areas, such as traffic congestion and environmental safety, is substantial.
Metromile is also built on an app. According to Preston, there is a huge amount of engagement with the app, which means customers are trusting them with their data. Some features that are included in the app might sound trite to some drivers, but with features like alerts for street-sweeping or snow removal, which can mean the difference between getting towed and staying parked in a major Metro area, Preston says that 70 percent of Metromile customers use the app frequently.
“It’s not something we want to force the customer to interact with,” he said. “It’s more like an end point. It’s helping you out. It’s alerting you when things are important, and it is a major factor in creating the kind of product experience that gets people to stick around and suggest it to friends.”
Metromile figures that it saves the average customer $741.00 per year. Preston pointed out that when someone signs up with Metromile, they don’t get any less coverage with the per-mile model, and people often upgrade their coverage because they save money. And because they are driving less and seeing the benefits, there is less risk. Metromile’s current customer portfolios are profitable, and the savings are then passed on to customers. Most of those customers stay under a 12,000-mile-per-year threshold.
According to Preston, Metromile has no set customer profile. The company’s early assumption was that it would attract millennials living in metro areas. While that part of the portfolio is starting to grow, they are also finding a much broader appeal. For example, recently retired customers like Metromile because their cars are basically used to run errands. Or maybe a customer who takes public transit to and from work will use the car to shuttle the kids to events.
Preston is passionate about the Metromile model, but he is most proud of his company’s performance when things go wrong. One of the basic driving forces for starting the company was that drivers were not getting a positive claims experience that was consistent with the engaging and simple branding that insurance companies were putting forward. For example, when a customer calls a claims adjuster, they often ask dozens of questions, but none that will really help the driver get back on the road quickly. Metromile basically calls on the sensor device to reconstruct the accident scene. It knows where the accident happened, how bad it was and which part of the car was damaged.
“So even the majority of claims are fully automated,” noted Preston. “On the app, the customer can enter just a few pieces of information and they get instant approval to go schedule a repair. They then bring the car to the repair shop, and that’s when they recognize that smart modeling of the data from the device is really a huge differentiator for them and for us.”
Preston has no illusions that big insurance companies will adopt the per-mile model anytime soon. They have a lot of catching up to do in terms of technology, and have not yet shown the ability to create a positive customer experience when they have to intervene in an accident.
“The big insurance companies don’t have the kind of technology necessary to make a lot of these interventions happen, nor do they have the expertise,” he noted. “But I expect that to evolve and change over time.”
We’re always on the lookout for opportunities to partner with innovators and disruptors.
Learn More