The topic of connected car data gets a lot of air time. While I’m completely sold on the possibilities, I spend my days unpacking the reality: connected car data is hairy. It’s high frequency, coming in from a multitude of streams, and it’s overlaid by rapidly evolving models for vehicle ownership and personal mobility. This hybridization of data creates a lot of questions and complexity. How do we capture, organize, and apply these newly available data streams? How do we distill this data to generate insights that can create differentiation?
Operationalizing hybrid data is the future for insurance.

Expanding Data Sets

The typical insurer’s existing infrastructure is not designed to process today’s massive-volume of data. The data coming into their core technologies right now is not typically coming from the wide range of devices in play: OBDII, mobile apps, and the vehicle itself, but we’re on the precipice of change, where they will need to digest and manage multiple simultaneous streams of data. For example, connected car data is growing exponentially. Depending upon the sophistication of the vehicle, there are 30 to 100 vehicle nodes (i.e., electronic controlling units or ECUs) per vehicle[i] that tie to more than 20 different suppliers, all of which are transmitting different data sets.[ii] Estimates suggest that a single connected car will send up to 25GBs of data to the cloud every hour.[iii] Let that sink in for a minute; 25GB of data is equivalent to 6,000 music downloads or 1,250 hours of browsing the Internet.

Another thing about most auto insurers’ existing infrastructures is that they can’t currently identify or assign values to the many different types of trips comprising the increasingly fractured vehicle ownership landscape (i.e., was the driver ‘on duty’ for a ride-sharing service, driving a shared vehicle, a named driver on a multi-car policy, etc.). These two factors can have a major impact on underwriting and risk management.

Related, but not entirely a technology issue, are the insights contained in these data streams. Technology absolutely plays a role in helping carriers distill information in ways that yield insights not previously available or applied. But, technology alone can’t solve the shift in mindset needed by many carriers to be open to accepting and applying the newly available data sets to evolve their traditional business models. Most auto insurers—both in personal and commercial lines—continue to rely on limited data sets that feed traditional business models. Since the first car insurance policy covering liability was issued in 1898 (to Truman J. Martin),[iv] followed by auto, fire and theft insurance becoming available in 1902 (the precursor to multi-line auto),[v] insurers have remained in this tradition; inclined to slowly evolve.

All Lines, All Drivers, All Vehicles

As all of this data is increasingly collected from a multitude of places and driver types, every line of auto insurance, every driver, and every vehicle will be impacted. This trend makes the operationalization of today’s hybrid data an absolute necessity for insurers; and being exceptional at it leads to new opportunities for growth. For example, there is an opportunity for carriers to achieve risk management success across a host of functional business areas by leveraging data insights and developing new, increasingly personalized underwriting models that take into account dynamic risk profiles. In other words, carriers will need to become good at using actual data on specific risks, while also mining and applying insights derived from new and existing models that build upon themselves to deliver improved ratings, policy personalization, and ultimately company differentiation.

Let’s look at a couple of examples.

Personal Lines Auto, Usage-Based Insurance (UBI). Because of advancements in connected car technologies, insurers with UBI programs can now connect with OEMs to access a digital history of a new vehicle, including automotive diagnostics, crash avoidance systems, and the ongoing capture of telematics data. Here, the connected car represents just one source with multiple streams of disparate data. But let’s assume this new vehicle is one of several vehicles on a family policy. Typically, a multi-car policyholder will have vehicles from different manufacturers or different model years, some connected ‘off the assembly line’ and some connected via an OBDII or the driver’s smartphone. The rich telematics data gathered, including velocity, turns, braking, weather, and road conditions, along with distracted driving information synchronized across the multiple methods of data collection, creates unique challenges, yes. But it most certainly creates opportunities if it is normalized and presented in a way that make sense.

Fractional Vehicle Ownership. Shared mobility as an alternative to privately owned vehicles is growing as a mobility model with one out of 10 cars sold by 2030 likely to be a shared vehicle.[vi] These shared ownership models, along with evolving opportunities with commercial vehicles and fleets, and the growing popularity of car-sharing programs, all add complexity to data capture and management. And the trend toward shared liability, where multiple insurers using telematics data to cover the multiple drivers of the vehicle based on who drives when, is a new frontier for insurance. This new frontier must be understood and addressed because today’s consumer is not backing down from ‘mobility my way’ for personal or even professional use, and operators of fleets need more transparency and control over the usage of their vehicles with a keen eye towards safety. We are looking to existing fleets as the first wave of opportunities to enhance the usage models of vehicles as well as create opportunities for insurers to develop new products in a business-driven marketplace.

What’s Next

As state regulators evaluate possible driving and vehicle ownership scenarios, and federal agencies examine data rights attached to the future of both personal and commercial lines coverage, one thing is clear: Insurers are at a crossroads in their efforts to seek and efficiently support these new business opportunities.

Dealing with either example/scenario above requires an infrastructure to handle the literally hundreds of variations these data sources present. A hyperscale platform can meet and scale with the processing demands presented by the vibrant data coming in as well as fuse the data with specialized software applications that can enact data by plugging it into the insurer’s existing workflow. This latter component is often overlooked; the most meaningful data left unapplied or unconnected to the business workflow isn’t all that useful. It needs to be operationalized if to enable improvements in overall risk modeling and assessment, as well as policyholder engagement programs.

Although some may tell you this is possible, the reality is that insurers are not quite there yet—at least without the help of a technology partner that can augment and normalize the insurer’s data to generate actionable insights that can be used to better understand risks and personalize policies, helping to drive customer loyalty.
Insurers that are taking action have a lot at stake. While these data streams create additional opportunities, careful attention must be paid to “getting it right the first time.” This means capturing underwriting value by being able to identify and garner profitable business and retaining desired policyholders quickly and efficiently.

Imagine insurers being able to advance their existing business models beyond just the asset (the vehicle), and on to the individual policyholders or drivers in order to accommodate new trends in fractional ownership. Or creating models from a host of sources of normalized data to inform and advance other, even customized or individualized opportunities for cross-selling policies within the vehicle/owner’s household, such as renters, homeowners or on-demand coverages. Operationalizing the hybrid of data coming your way lays the groundwork to achieve and successfully act upon lifetime policyholder value. Sorting this all out isn’t insignificant, but it is required.

Mac Fraser, GM, Telematics, CCC Information Services Inc.

[i] Keeping, Steven. “How the Connected Car Will Disrupt Personal Transport.”
[ii] McKinsey & Company. “Shifting Gears in Cyber Security for Connected Cars.” February 2017.
[iii] McKinsey & Company. “What’s Driving the Connected Car.” September 2014.
[iv] Onge, Kenneth. “First Auto Policy Sold 110 Years Ago Today.” Insurance Journal. February 2008.
[v] Allstate Insurance. Archives.
[vi] McKinsey & Company. “Disruptive Trends that will Transform the Auto Industry.” January 2016.

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