Des Moines-based Homesteaders Life is a century-old funeral and end-of-life costs insurer that is applying technology to market additional services to its policyholders. Nick Gerhart, executive vice president at the company, spoke to Digital Insurance about the company's efforts, as well as insurtech startups, AI and regulation. Gerhart draws from his experiences as Iowa insurance commissioner from 2013 to 2017, as an advisor and board member at several insurtech companies, and other executive roles with Farm Bureau Financial Services and American Equity.
This article is from a longer interview and edited for clarity.
What is the ecosystem that Homesteaders Life is building?
It's 70% complete. We have a big goal, because no one's doing it very well right now. We're building up resources. The goal is to help funeral homes and their brand in the community. Typically, they're really good at the service, and then afterward, they don't really follow up as well. We have built solutions for them to stay in contact with a family for 12 months afterwards. We're trying to help them amplify their brand and their place in the community.
What are the pain points?
Most of it's user adoption. A lot of these are third- and fourth-generation businesses. Like any small business, they deal with what's in front of them. They don't really have the time to do much else. Our vision is, what can we take off their plate to add value for those families? And then really make it seamless.
What are the goals for Homesteaders Life's ecosystem?
To have the most cohesive ecosystem for an end-of-life solution, beginning at retirement or age 58 to death, and then back in to help that family through grief, or an estate settlement, other things they may need. Can we put these folks down a journey where they want to go and help them find that person, to help these families do a lot of things that need to be done over this entire continuum? You have all these jobs you have to do that most people don't want to do or don't have time to do. It could be 500, 600 hours or more to settle an estate. Most people have lives and jobs and they don't know where to go.
We're trying to build an ecosystem that can be national and digital. Then a local platform that would be able to connect them to people locally. If the family wants someone face-to-face to talk to, they should do that. If they think they can do it on their own, they should do that. If they start on their own, and have to hit eject, they should be able to do that.
What's your view of the current insurtech startup landscape?
I don't discount what [insurtechs] bring to the ecosystem of insurance. Because the consumer we're dealing with now and the consumer of the future, they're going to want an Amazon-like experience. You can't put them in a 1980s Kmart. Now it's a regulated business, so it's not like buying a pair of shoes. You're buying homeowners and life insurance. We have to figure out ways to transparently take down the friction, make it a better buying experience.
I don't believe agents are going anywhere anytime soon. They're going to be around for a while. For the next decade, agents are pretty solid. How can you help them amplify their brand, take stuff off their plates, serve more families and more customers? The providers that can figure out how to do that have a really successful business.
You're seeing much more collaboration between these providers and carriers. I've invested in a couple startups over the years. You have to be careful not to get stuck in what I call the insurtech petting zoo, where you're just a nice-to-have. You've got to get to the point where you have a real product, a real idea, real value for that carrier. And don't be afraid to push. Sometimes it'll take longer than it should. Most of these folks are on limited capital, and they have to be really smart in how they go after their customers. You have to find the right product, right market, go-to-market strategy – and how are you going to really attract those carriers, to get guys like me to say, Yeah, I'd look at that. That's so hard because we're busy.
In P&C, how do you inform policyholders about what they're covered for or not?
This industry is ready-built for AI to come in and solve some of these problems. You can take someone down a journey, and then you can say, hey, load your policy up. You can do OCR [optical character recognition] but you could really do an AI learner model, put your policy in, then run it against other policies in your neighborhood or your zip code.
The data is there. It's contract language. There are pretty cool tools that can compare policies. It has to be really simple. It has to be like, hey, click here and in three minutes, let me tell you what you need in this policy or what you're missing in this policy. AI was built for that. That would be a great business. You could do a lot with something like that, because everybody has this problem, even if they don't know they have this problem.
You might think you have the right coverage, and maybe your agent gave you the right coverage. They probably did, but it'd be good to know. Consumers need help, whether it's retirement planning or insurance planning. We've made it really hard on people, especially people that really don't want to think about it and just hand it off to someone.
What are regulators' challenges with AI and proprietary risk modelling?
In the regulatory world, AI could be a four-letter word pretty quick, if we're not careful. Most regulators are going to focus on bias, whether express or implied. You must have auditable, traceable models that people understand.
I'm a big believer in the state-based regulatory system, but it was not built to withstand huge catastrophic failures. You shouldn't be using these tools in a nefarious way where people are getting taken advantage of, or you're slotting people in a bad price because they don't know any better. Somebody has to tell [regulators] what happened and why. There has to be at least one person that understands what the model is.
There's ways you can make it confidential and protected, but there are some [days] that's a problem. I understand that if you poured a lot of money, a lot of data and you built this superior model, that's valuable, but you can't then expect a regulator, who has this dual mandate, to say I'll just trust you. The regulator still has to do their job to make sure that consumers in their market are being protected and getting the deal they're supposed to get, not having claims basically not paid.
