"The best loss is the one that never happens."
That, to me, is the most compelling line from the interview I conducted recently with Pete Miller, CEO of The Institutes. The Institutes (with which ITL is affiliated) has for some months now been arguing that insurers should take all the data that they're gathering on potential hazards in commercial buildings, homes, cars, etc. and spin it forward. Rather than view data analysis as a historical exercise for pricing risks, then indemnify customers for losses, Pete says — and I wholeheartedly agree — that there is a major opportunity for insurers to use sensor data to spot wiring problems before they lead to fires, to detect a leak before it can do any damage, to warn workers that they're risking injury before they get hurt.
Insurers will pay fewer claims. They will also avoid all the costs of processing the claims that don't happen. They will have the opportunity to sell all sorts of new products and services that minimize risk. Meanwhile, insurers will not only spare customers from paying a deductible but will let them skip all the hassle that comes with recovering from a fire, leak, etc.
Talk about fulfilling the insurance industry's noble purpose. Talk about creating loyal customers.
If you want to dig deeply into "predict and prevent," I encourage you to check out the podcasts Pete has been conducting with insurtechs such as Whisker Labs and Betterview, with executives from insurers such as Chubb and brokers such as Marsh McLennan, with insurance commissioners and with customers. (If, like me, you're more a reader than a listener, you can find the transcripts for each of the six podcasts on the same page.)
In the meantime, here is the interview I did with Pete, which, in my not so humble opinion, is a great summary of the opportunity that predict and prevent presents:
ITL:
The predict and prevent theme really resonates with me. How did the idea take shape, and how did it become a major theme for The Institutes?
Pete Miller:
The idea came from an awareness of emerging technologies that can really help to do more effective and, in fact, more-real-time risk mitigation.
If you think about it, the best loss is the one that never happens. And there's a huge amount of data, including real-time data from IoT sensors, as well as a vast increase in the tools that are available to process and make sense of the data. If you start to combine all that, then there's an opportunity for real-time, improved data analytics for resiliency.
As a result, we can make people's lives better and safer and easier. We can use analytics to predict what's going to happen and stop problems before they start.
ITL:
What are some of the best examples of predict and prevent that you've come across since you started focusing on this theme?
Pete Miller:
Whisker Labs has a device called a Ting that is really, really cool. You just plug it into a wall socket, and it monitors for electrical hazards both in your wiring and the electrical devices you're using and alerts you before a fire can start. State Farm is deploying Ting to its customers [at no cost to them].
There's also a device that people in a warehouse can wear. It learns your body for the first few minutes. If you’re taller, you do some things differently than if you’re shorter, for instance. After that initial phase, if you're doing something like leaning over too far and are risking injury, the device goes, "No, don't do that."
How cool is that? You prevent injuries that could have someone walking around in all kinds of pain for a lifetime.
Others are looking at ways to head off damage from wildfire.
It's all been really interesting.
ITL:
You have lots of senior insurance executives on The Institutes board and talk to others all the time. What has their reaction been to predict and prevent?
Pete Miller:
I can't talk about actual deliberations within our board, but I can say that there's a very broad recognition that this is the right thing to do. The theme really resonates. It's just too logical, and if the insurance industry doesn’t do this then somebody else will. There's so much benefit to be had.
Insurance companies understand that a predict and prevent approach cuts their losses. They don't just have fewer claims but also avoid all the expenses associated with loss adjustment. The loss never happened.
Policyholders are happy because you're telling them you can detect problems and prevent losses.
Regulators see this as a great opportunity, too.
ITL:
I imagine that there are also ways to take the data being gathered and feed it back into building codes, insurance contracts or other instruments that will change behavior.
Pete Miller:
There are a variety of ways, some of which is dual-purpose. Chubb, for instance, is working with a real estate company that has a bunch of office buildings, and they put water sensors in them. The sensors provide real-time alerts if there is a water leak or flooding, and a valve can shut off the water automatically. The sensors also generate data that can be provided to builders or property managers or those who write the building codes and that will help head off similar problems down the road.
ITL:
To me, the tipping point for predict and prevent has always depended on being able to make a strong economic argument. If you retrofit a house to install sensors and a valve that can automatically shut off the water when there’s a leak, that costs you maybe $700 or $800, and you have to install a lot of those valves to prevent a single loss. How close do you think we are to a tipping point?
Pete Miller:
In round numbers, the cost to set up and monitor those sensors that Chubb installed is less than $5,000. A single water loss? $100,000, on average. So, for that client, the math is easy.
Whisker Labs says its Ting device can prevent 75% to 80% of house fires, and the cost of a device is minuscule.
Those economics are coming into line. It's starting to make sense. And the costs are always coming down.
Companies are also viewing predict and prevent as a value-add in terms of customer satisfaction. Customers think, "I have an older home, and my insurance company is giving me a device that protects me and my family." That's worth something.
ITL:
I saw in your podcast interview with the founder of Whisker Labs that he also talked about how devices like his Ting get smarter as more get deployed. You can see, in his case, that a certain fluctuation in the flow of electricity in a building that didn’t initially seem significant actually shows a risk of fire.
How do regulators deal with this new approach? Do insurers just graft it onto the existing insurance model? Or does there have to be a more fundamental change?
Pete Miller:
That's a very good question.
Insurers understand that this is a cultural shift, and, frankly, there's some resistance. Some people say, "Well, that's not insurance." But the people I've talked to say, "Well, now we're kind of thinking we're risk managers." Insurers will certainly always indemnify clients for losses, but if we think in terms of risk managers then there's a whole series of broader products and services we can offer.
There are other risks that can be mitigated, too, beyond fire and water. Cyber, for one. Insurers can help clients reduce those risks and eliminate a lot of losses. On the commercial side, there are lots of additional property risks that insurers could help reduce. More can be done with telematics in cars.
And I keep going back to the selling point: You, the insurer, proactively stopped something bad from happening.
You didn't just save me my deductible. I don’t want to have to do a bunch of stuff to recover after a loss. If I own a business, I don’t want to have down time.
It's the same reason I go to the dentist, right? Because I don't want all my teeth pulled.
You asked about regulators. Every regulator I talk to says this is awesome. They look at predict and prevent from a consumer point of view, and they say, "You just saved the consumer a lot of heartache."
ITL:
Words to live by.
And it sounds like the opportunities will only grow from here.
Thanks, Pete.
Cheers,
Paul