Usage Based Insurance and Big Data – What is a Carrier to Do?

sma ubi tableThere is little doubt that Usage Based Insurance (UBI) (a.k.a. Telematics) is a hot topic in the U.S. Insurance Market. A recent survey from Strategy Meets Action found that while only 18 P&C insurers have an active UBI program in more than 1 state, 70% of insurers surveyed are in some stage of planning, piloting, or implementing UBI programs.

A carrier cannot venture into this space without considering the data implications. Usage Based Insurance, whatever its flavor, involves placing a device in a vehicle and recording information about driving behavior. Typical data points collected include: vehicle identifier, time of day, acceleration, deceleration (i.e. braking), cornering, location, and miles driven. This data can then be paired with publicly available data to identify road type and weather conditions.

Now consider, a 20 mile morning commute to work that takes the driver 35 minutes. If the data points noted above (9) are collected every minute, that 20 mile commute would generate 315 data points (about 16 data points per mile driven). If the average vehicle is driven 1000 miles in a month, it would generate 16,000 data points each month or 192,000 data points each year. Now consider what happens if a carrier enrolls even 1000 vehicles in a pilot UBI program. Within a year, the carrier must accommodate the transmission and storage of over 190 million data points. Progressive Insurance, the leader in UBI in the U.S. market, has been gathering data for 15 years and has collected over 5 Billion miles of driving data.

Even more critically, the carrier must find a way to interpret and derive meaningful information from this raw driving data. The UBI device won’t magically spit out a result that tells the carrier whether the driving behavior is risky or not. The carrier must take this raw data and develop a model that will allow the carrier to score the driving behavior in some way. That score can then be applied within rating algorithms to reward drivers who demonstrate safe driving behaviors. As with all modeling exercises, the more data used to construct the model, the more reliable the results.

While data transmission and storage costs are relatively inexpensive, these are still daunting numbers, especially for small and mid-sized carriers. How can they embrace the changes that UBI is bringing to the market?

From a pragmatic perspective, these smaller carriers will need to partner with experts in data management and predictive modeling. They will need to leverage external expertise to help them successfully gather and integrate UBI data into their organizations’ decision making processes.

In the longer term, credible 3rd party solutions are likely to emerge, allowing a carrier to purchase an individual’s driving score in much the same way that credit score is purchased today. Until then, carriers need to make smart investments, leveraging the capabilities of trusted partners to allow them to keep pace with market changes.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s