Why AI is the future of underwriting
Todd Rissel, chief executive officer, e2Value
As part of Intelligent Insurer’s Underwriting Innovation USA online event, Todd Rissel of e2Value will be explaining how to augment traditional rating techniques with real-time insights to support timelier and more accurate asset valuation. Here he explains how e2Value is making waves in this arena.
Data is key for the re/insurance industry—but how reliable is that data? In a fireside chat as part of Intelligent Insurer’s Underwriting Innovation USA online conference, Todd Rissel, chief executive officer of e2Value, will offer his take on the data challenges faced by the industry—and the solution his company has created.
As a provider of web-based property valuation solutions using artificial intelligence (AI), e2Value sets out to make the valuation process faster, more accurate and easier to use, augmenting traditional rating techniques with real-time insights.
“The data can be gathered, cleansed, checked, verified, and put to bed with the sending of a text.”
Rissel’s fireside chat will address how detailed structural characteristics of individual properties can be incorporated into the risk assessment process for a more accurate appraisal of risk; how to maximise the data culled from predictive modelling to improve the accuracy of underwriting and pricing; and how to take the inconvenience and expense out of inspections with quick, unobtrusive and contactless data-gathering.
Here he explains how he arrived at his solution.
Tell us how and why you came to establish e2Value.
We came to London a few years ago and in meeting after meeting underwriters and brokers asked for data. Everyone wanted data about a structure. The problem seemed to be that “data” was an imprecise term. At the time, any data was considered good—but the structure data was not perfect. Imprecision created doubt and the available data created imprecision.
What issues were underlying this problem?
On one end of a transaction is a risk modeller. Risk models need certain sets of data in a strict format. It is meant for the modeller’s use. The model looks at what’s there and what natural or manmade disasters can do to such a structure. The model has nothing to do with replacement cost, or the proper value for the structure, or what will be replaced.
On the other end is the broker, client, carrier, and valuation. After the loss, the broker, the client, and the carrier care about covering the financial impact of the structure. If we know what is there today, we can calculate what it will be replaced with and the cost of that replacement.
So—it turned out every structure has two parties with varied interests and two sets of data and two outcomes for very different audiences. Also, data meant for the realty market or to raise taxes may have different phrases and objectives than insurance, as well as a different definition of timely. It may take two years to update a tax database and therefore, an underwriting application.
But the client may need a builder’s risk policy now or have just bought a newly rebuilt home. They don’t have two years.
The problem was that it wasn’t enough just to get data; it was getting timely data everyone can trust and then use for a particular sector of a particular market.
What did you do to solve that?
We moved valuation from a process that was two-dimensional and easy to hack, break and falsify, to a three-dimensional, thinking, AI system. The result was a unique, now patented, process for valuation.
Its core centres around breaking structures down to a standardised, consistent set of data and matching that to each zip (and postal) code in North America. With just an address we can value a one-room cabin to the most luxurious and expensive homes in the world, and everything in between.
Pronto is e2Value’s main service. It uses publicly available data or client-supplied data to deliver a value and the related data. It can be used for any home or commercial building and is ideal for lead generation, initial quotes, field services and claims.
What benefits has this created for the industry?
Most portfolios, not reviewed by e2Value, are over-covering smaller risks and way under-covering larger risks. Carriers use us to help them set the base as to data and values. Then, they can use a scalpel to apply rate changes where needed, versus a blanket approach.
Our solution is a timely way to have the data checked, updated, if needed, imagery added for verification, all in an easy-to-use, easy-to-integrate process—even a process where you can gather the data now and get a download of all the data later.
The data can be gathered, cleansed, checked, verified, and put to bed with the sending of a text. The key is getting data in a way that is easy to use, update, change with systems and technologies, and verify.
What other new possibilities does this create for the industry?
If you’ve never had a process to engage brokers in data-gathering, now you do. You’ve never engaged your end client? Now you can.
How has the industry responded?
We now have a wide range of customers from London to Canada and the US, from one-person brokers to top 10 carriers.
We tend to be used first for the “difficult” risks, such as high value residential, manufactured/mobile homes, farm structures and then we are then brought into mains street lines and regular commercial.
What lies ahead?
I now expect a move to more actual, real-time, on-the-spot data-gathering using everything from satellite imagery to internet of things data.
In the end, I believe, it needs a simple way to measure, underwrite, and rate the risk. That’s where we are excited about how we can help clients do just that.
Todd Rissel is one of the speakers at Intelligent Insurer’s Underwriting Innovation USA virtual event addressing how to accelerate underwriting profitability in a world of changing risks, taking place between November 10 and November 12, 2020. Full details can be found here
e2Value, Underwriting, Data, Insurance, Reinsurance, Todd Rissel, North America