In a recent article The Economist described how new technology allows insurers not only to pool and insure risk, but also provide incentives for customers to reduce risk exposure by modifying their behavior. One insurance company for example, equips customer cars with monitors and tells them when they drive unsafely and how. Those receiving such insights crash less. Another company, which insures people with manageable diseases such as HIV and diabetes, offers them free monthly check-ups. If these show that they are not sticking to their treatment regimen, their premiums go up. People respond by looking after themselves better rather than pay more. Its policyholders went to hospital less and reduced their medical expenses after joining its health-monitoring scheme.
A report by Morgan Stanley and the Boston Consulting group predicts that damage to insured homes will fall by 40-60% if all the latest technology is adopted. The risk pools for home and car insurance will shrink. Insurers using the latest techniques to cherry-pick the best drivers would receive just a ninth of the claims of firms insuring in conventional ways.
Leaving aside problems that such developments might create (fundamental shifts in risk pools might reduce risks and premiums for some, while leaving others uninsurable), a couple of things are particularly interesting:
First, insurers are expanding their business models from simply providing risk cover to engaging customers to co-create value through behavior modification. Like a servitization textbook example, the insurance experience becomes less “transactional” and more “relationship based”. By reducing deadweight costs of uncertainty (through behavior induced risk reduction), premiums can come down benefiting customers, while insurance margins increase (through reduced claims) benefiting vendors. Nevertheless what is good for individual innovating insurers might not be good for the industry as a whole. As net risk exposure declines, demand for the core insurance product may reduce. Furthermore it seems that the core competence in the business is shifting from actuarial risk calculations to big data based predictions of behavior / outcome correlations and design of behavior modification inducement offerings. Accordingly big data firms such as Google or Amazon might be expected to consider this market –as they are in effect doing- posing a considerable threat to established players.
Second, how this translates into industrial environments:
- Initially insurance premiums for operating industrial products and assets could be much more closely set to track asset condition predictions and maintenance practices and behaviors which will be made much more transparent and granular through the IoT and big data analytics across large asset populations: Better practices would be rewarded with reduced insurance premiums (value co-creation)
- Then OEMs would probably come under pressure to expand “trouble free operations” promises, which will lead to increased inherent (designed in) reliability of products (already evident in the automotive and other industries)
- In turn this will mean shifts in service revenue flows –in essence customers will change their default position and pay to avoid failure rather than to have something repaired. However these most probably will not be additional payments, but already embedded in the product price –which will not increase, as this does not reflect additional value but reduced cost.
New technology creates the potential for reducing the cost of achieving outcomes –mainly by eliminating deadweight costs, in this case the costs of uncertainty regarding failure, needed to be covered by insurance. Data rich and analytics savvy manufacturers may benefit, either by effectively insuring customers themselves for competitive advantage and/or by offering (monetizing) data and insights to others to provide the service.
Beyond asset management, reducing costs of uncertainty has implications along entire supply chains and business processes, as statistical tools familiar from TQM (zero defects) and predictive algorithms can be applied across orders of magnitude greater number of data points. Capability to reduce operational risk (better than others) can have significant competitive advantages along different dimensions (costs, inventories, quality, time-to-market, throughput and cycle times, project execution, design of offerings, pricing…).
To-date the focus in big data analytics has been mainly on analyzing customer behaviors and marketing. However the competitive advantages through improved operations and servitisation should not be underestimated. Platforms such as GE’s Predix and others will be powerful tools in this era of data driven competition -not only narrowly by improving availability and reducing operational cost but more broadly by providing genuine potential sources of competitive advantage.