Have you met the insurance industry yet? The answer, for crowdsourcers, is a resounding, “Yes.” They’ve met and had coffee, and now the relationship appears to be blossoming into something deeper.
The introduction happened in 2001, when Allstate launched a successful Kaggle campaign called “The Claim Prediction Challenge.” More than 1,290 entries were submitted by expert statisticians attempting to create an algorithm that could successfully predict injury liability based solely on the characteristics of an insured vehicle.
The company gave entrants three years’ worth of data on insured vehicles and the injury claims associated with them. Each team worked to find a link between each vehicle and the payments made on their claims using metrics ranging from the number of cylinders in the engine to the actual length of the car. After three months, a winner emerged, with an algorithm 271% more accurate than Allstate’s previous model for predicting claims. Today, the carrier uses this crowdsourced solution in its product and pricing strategies.
But things haven’t stopped there. Some insurance giants want to use crowdsourcing to tap into a strategy that could change the way the industry regulates rates. That’s a big idea.
Changing the way people pay for car insurance
If you’re not an insurance expert, the term “telematic usage-based insurance” may sound like coverage for Doc Brown’s flux capacitor. But it’s not really that far out there: It’s a new type of coverage that offers dynamic rates based on changes in risk – and it’s gaining some traction by crowdsourcing driver data.
Companies such as Progressive, Liberty Mutual and Allstate have developed telematics devices – a device you plug into your car that transmits real-time data about policyholders’ driving practices. It tracks mileage, speed, time-of-day information, driving actions (such as sudden stops) and much more. With this technology, drivers and insurance providers operate within an immediate feedback loop that offers more accurate rates based on driver risk.
For example, if a policyholder who once commuted to work every day suddenly starts taking public transportation instead, his insurance company could immediately reflect that reduction of risk – driving fewer miles cuts the chances of an accident – by lowering his insurance rates that month. In an age of instant gratification, rewards such as this could prove to be a powerful incentive for drivers to adopt safer driving habits.
Changing behavior behind the wheel
A car crash occurs in the U.S. every 4.8 seconds, and the Insurance Information Institute reports the average cost per injury claim to be close to $15,000. So, what’s the best way for insurance giants to test the effectiveness of reducing these figures through telematics systems? With a crowd.
A large number of top providers have begun collecting driving data from employees, corporate fleet vehicles, interested policyholders and other sources. Most recently, Allstate launched a crowdsourcing campaign to test its Drive Wise system among a large group of employees and agency owners. Based on more than 11 million miles and 350,000 hours of driving data, it found that telematics usage-based insurance has great potential to change behavior behind the wheel:
At the beginning of the test, only 25% of participants scored in the ideal driving “safe zone” (rated by speed, sudden stops and more). By the end of the test, the number of drivers practicing safer activities behind the wheel rose to 75%.
Insurance and crowdsourcing in the future
What does this mean for the future of insurance? For one, telematics will put drivers in control of their own car insurance rates – the safer you drive, the more savings you earn. Plus, crowdsourced data seems to indicate that these systems will likely play a huge role in helping to reduce accidents, injuries and property damage claims – which translates into lower insurance costs across the board.
In return, the insurance industry could help transform the way the public thinks of crowdsourcing. It’s bigger than problem-solving in a group of experts or a way to fund a Veronica Mars movie. With tests and patents also on the way for detecting distracted driving, drunken driving and even texting while driving, crowdsourcing could play an integral role in saving lives, preventing accidents and cutting costs. Not bad for a relationship that’s still in its early stages.
This article was contributed by Kelly McMurtrie, a writer for HomeInsurance.com. Kelly has been writing content for HomeInsurance.com and other major brands since 2011 after graduating from the University of South Carolina with a B.A. in Media Arts.