Q&A: Ryan Kottenstette, CEO, Cape Analytics

Over the last year, we’ve tracked a number of new insurance tech startups in the homeowners and renters insurance arena, but there are also some taking aim at property data incumbents in the insurance industry with new technology.

Last October, Cape Analytics announced it had raised $14M in Series A funding led by Formation 8 to use computer vision and machine learning on geospatial imagery to generate structured property data for insurers and reinsurers. This week, Cape Analytics announced a partnership with NearMap to give Cape full access to NearMap’s US aerial imagery library.

We chatted with CEO Ryan Kottenstette on why co-founded started Cape Analytics, the future of property inspections, and whether a possible bifurcation along tech adoption is coming to insurers.

What was the impetus for starting Cape Analytics?

At Khosla Ventures, I was lucky enough to work with a crop of interesting startups across verticals but, in particular, worked with a handful of startups for which the fundamental technical innovation was applying deep learning and machine learning on imagery to extract data. So I was lucky to have a front row seat and play a proactive part in incubating and building some of those companies.

On the other side of the spectrum, Khosla Ventures had invested in Skybox Imaging, a satellite imaging company that proved to be very successful, and we looked at a whole bunch of satellite and imaging drone platforms. To me, there were two very clear technology inflection points happening: 1/ the rapid advances coming in deep learning and machine learning on imagery and 2/ a massive proliferation of access to space imagery as coverage and frequency increased, while costs decreased. So the natural thought progression was: Let’s take these two technology inflection points, which often creates the cracks for which new innovative opportunities can be created and then let’s start looking at industry verticals to try and understand where there are clear market-driven needs that can be solved with these two major inflection points.

And so, I spent a handful of time with my co-founder looking very clearly at a whole bunch of different verticals and the net sum of that exploration was that this is not the type of change where there is a relatively narrow set of opportunities, There are a lot of valuable opportunities, so the hard part for us was to stay disciplined and not try to be everything to everyone. So right now we’re exclusively focused on property insurance and reinsurance as an industry vertical for Cape Analytics and that’s allowed us to be very receptive to really understanding what our customer needs are executing those things well.

Given how technology is advancing and getting adopted by the insurance industry, is there a situation in which some players get left behind? Do you foresee a potential bifurcation in the industry?

I’m not sure how it’s going to play out. On the one hand, you could think about a forthcoming bifurcation. But what is that bifurcation?

Over the last 3 to 4 years, there’s been a lot excitement around insurance tech and that’s really been an awakening of the insurance industry saying, “We’ve not innovated as fast as the generational consumer expectations have changed, technology tools have come available and those of us who don’t innovate are going to get left behind.”

I think there’s absolutely going to be a bifurcation between insurance companies who are doing a very good job of being data-driven and leveraging today’s data and toolsets and those which aren’t embracing them. But as more technology providers built on new-generation toolsets come to market, that’s also an opportunity to democratize property insurance in that even small insurance companies can make good use of these tools. So I think technology adoption is going to be the demarcation line, not the size of the incumbent per se.

On partnering with aerial imagery providers

We have a long-term and strategic partnership in place with NearMap that basically allows us full access to all their imagery across the continental US and the full frequency of updates. It helps they have a great product with a very valuable mix of coverage accuracy and frequency especially for the market verticals we’re targeting. That said, we have multiple key partnerships in place; this happens to be the first one we’ve announced publicly.

There’s a lot of good businesses surrounding image capture, but we view our primary value-add as targeting customers and helping customers who maybe don’t want to put the time and resources behind manually reviewing imagery and want answers at scale in an automated way. So we view NearMap certainly as a critical partner, but by no means a sole partner and we’re lucky to have a couple of major image partners of high quality and good coverage that we’re working very closely with.

On the future of property inspection

We’re absolutely not trying to bill ourselves as a replacement to property inspections. I think our offering is a lot broader that that simple economic measure. We have some clients who use us to be more discerning for which property inspections they order, but we have others who haven’t changed their inspection volume at all and instead use us for a couple things. The first is that having the data instantly means they can take fewer actions on the inspection. For example, they don’t have to reprice the policy after they bind it. The second thing is more intelligently determining which properties they want to inspect.

In general, the residential homeowners underwriters, they’ll vet somewhere between 0 and 100% of new risks. They basically don’t inspect renewal business at all or, if they do, it’s maybe on a three or five-year cycle. What we can do is give some intelligence into what properties they want to inspect. So for a piece of renewal business that built an extension or a guest house or pool, they might want to go check that out.

Looking for insurance tech startup data and analytics? Sign up free for the CB Insights Venture Capital Database.