Does leveraging driver data justify the cost of telematics?
The True Cost of Telematics
Many insurance companies indeed turn to telematics because they want to improve pricing and stay competitive with industry leaders. The thing is, adding the cost of telematics increases the overall expense ratio slightly. But if you stop reading there, you’ll miss the bigger picture:
The potential profit of leveraging telematics offsets the cost increase
In other words, the expense might go up a bit, but profits will likely go up as well, and the potential for these profits may greatly outweigh the additional investment. What’s more, with the increasing use of mobile apps, the cost of telematics is dropping dramatically. Also, as insurers place more reliance on telematics data in pricing, the value of telematics will increase, and the value and cost of proxy variables will decrease.
Let’s look at both of these scenarios more closely:
- Profit Offsetting Expenses
- Replacing Current Proxies
Potential Profit of Leveraging Telematics to Offset the Expense
Target and scale: these are the two words that come to mind when I think of how to effectively offset — and even outweigh — the cost of telematics.
Many carriers employ discounts as a blunt tool. But with telematics, you’re able to be more strategic and purposeful about effective discounting. In other words, you can target your discounts. Targeting, that provides discounts for specific groups of better drivers, will set you up for better pricing and margins.
Scale is also important. Until you have scale, it’s almost impossible to get traction in telematics, and when you’re starting, you don’t have scale. The best advice I can give to insurance companies is to farm out the areas where they lack sufficient scale and that are not their strength. A technology partner can set up a telematics program for you more quickly, efficiently, and effectively. If that partner also has the right data and in large scale, you can immediately catch up to the competition on pricing with some assurance. This is critical.
As more and more carriers move to telematics models, the newer business will be driven by this kind of technology. Behavior-based pricing appeals to younger people because of their comfort with the associated technology and process. If you’re not attracting the younger generation at scale, you’ll end up sitting on an aging group of clients.
I’ve worked with several of the top ten insurance carriers who do it right and they obviously have no problem offsetting the cost of telematics, and becoming more profitable in the process.
Discovering Better Predictive Variables (to Replace the Old Proxies)
Using credit variables as a proxy for driving behavior has grown tremendously in the insurance industry. It turns out to be a fairly good predictor of claims risk. The more favorable the credit variables, the lower the risk. But it’s not always accurate. Some people with favorable credit histories may turn out to be highly risky drivers and some people with less favorable credit histories might be excellent drivers. On average, credit improves segmentation, but it is still a blunt proxy for actual driving behavior and risk.
To help minimize these outliers and hedge your bets, as an insurance company, you combine other proxies: demographics, for example. Age, where you live, what you drive, how far you drive, marital status, and other factors get averaged in to help predict loss potential more closely. Yet a youthful male doesn’t need to be surcharged simply because he’s a youthful male, and consumers understand this.
No matter how many proxies you add to the mix, they are still proxies that are correlated with driving behavior and claims risk, not causation of loss. And even predictive variables that are easy to collect still aren’t free.
This makes me think of the insurance application that a company I worked for utilized years ago. It was a long application, filled with all the data points we wanted to gather.
When the company switched from having agents assist with the application to having consumers fill it out directly, we realized we were losing potential insureds before they got to the competitive rate. We were asking too many questions.
So, we shortened the application by ranking all of the questions according to the segmentation value provided and the time required for the customer to answer accurately. It essential to get consumers through the process quickly and the marginal gain of several questions was simply not worth the time. Collecting too much demographic data was costing us potential high-value insureds and driving up the average acquisition cost per completed quote.
In the future of staying profitable, insurance companies must collect the data to create variables that reflect real driving behaviors. But how will you do that? Driving behaviors will serve as more advanced predictive variables, replacing old proxies. It’s the only way to remain competitive in the industry.
An Ideal World of Rating Risk
In the future of rating risk, insurance carriers will leverage real mobility and driver behavior data to determine variables that reflect that real world. And to do that, you need to attract the new business and be in a position to capture, validate, and derive insights from the data.
The insurance industry is not there yet, but as a whole, it’s working toward three things:
- Telematics at point of sale: This will help build better pricing from the start.
- Consumer participation: When consumers understand that they can influence the price of their policy, behaviors may change. They’ll seek out programs where they have that control and potentially drive more safely to save money.
- An interactive model: Insurance companies tend to price everything else then add a discount for telematics at the end, which weakens the power of prediction.
Telematics at point of sale and consumer participation is fairly straightforward concepts. Let’s look more deeply at the potential of an interactive model.
The Interactive Model of Telematics: Multivariate Analysis Is Key
When insurers originally introduced credit as a variable, they used it to tier insureds, then applied multivariate models to each of the resulting tiers. They quickly moved to full multi-variate models that incorporated credit variables, which significantly increased their predictive power.
It has taken longer to build multivariate models with telematics because insurance companies have lacked penetration of telematics among insurers.
Multivariate models leverage real-time mobility data, which can change, compared to traditional predictive variables, which don’t change that frequently.
A telematics program that leverages risk across different modes of transportation and has collected the evolving data for some time is the perfect resource for underwriting interactive models that will keep an insurance company competitive.
Start Here: Three Fundamental Elements to Be Successful with Telematics
When considering investing in new programs, such as telematics, to improve business, you must have these elements:
- Technology: the app, the algorithms, the collection methods
- Data: the right type, quantity, and quality
- Expertise: the know-how to build internal processes and quoting that are unique to a telematics product and an understanding of insurance (if you are outsourcing)
Telematics is not new. We don’t need to learn how to do it from scratch, and we don’t need to pilot a program to figure out if it works. Start with what’s already known to get up and running fast. If you have the right expertise, data, and technology, no pilot is necessary.
It’s not “how much does it cost?” or “how much will we save?” It’s how the whole process will work together for an experience that’s beneficial to both the insurance company and the consumer in the long run. In my experience, telematics is proven to work.