Dom Sindall & Stephen Sheehan: Marsh Re: From days to minutes: rethinking facultative reinsurance (417)

0:08Stephen Sheehan: Dom, Stephen, great to have you. You are respectively Head of Digital Broking and Head of Analytics for one of the world's largest reinsurance brokers, Marsh Reade, formerly known as Guy Carpenter or Guy Carpenter Facultative Reinsurance in your case. Welcome.

0:22Dom Sindall: Thank you, Matthew. It's a pleasure to be on the show today.

0:26Matthew Grant: Yeah, thank you, Matthew. Really excited to speak with you today.

0:28Stephen Sheehan: Well, I'm also very excited to really properly understand facultative insurance or facultative reinsurance, I should say, because it's something we all hear about. But actually, when we try to understand who the client is and what problem you're solving for them, it is quite important to unpack that a little bit. So does one of you want to sort of have a crack at talking about, first of all, who are your clients? And then what is the problem you're solving for them in the business? And then I'll get us on to the solution you're offering.

0:54Dom Sindall: I'm glad you're fascinated. Our clients, being Marshree, are the insurance companies, ultimately. And when we talk about facultative reinsurance, we're talking about insurance of insurance. So I'm going to pretend for a minute you are, you know, a multimillionaire. You might say you own 10 condominiums in North America. They're worth$100 million in terms of its insured value. What do you do when you need to go and get that insured? you're going to go to your retail broker and ask them to kind of sort coverage for you and when that information and that submission as we would call it kind of gets received by the retail broker they're going to go into the insurance market and seek out seek out coverage now if i just for the sake of this example give you two possible outcomes now if i go down the first scenario as a retail broker if you approach the insurance market and they do not have capacity or appetite to write condominiums more than say$50 million, then they're just going to decline your business opportunity and not insure you.

1:56Dom Sindall: And that means the retail broker has to go on to somebody else and they find another insurance company who could maybe do that. Now, the alternative is that retail broker approaches the insurance company and that insurance company goes, well, I know I can only retain$50 million, but they then go into the facultative reinsurance market and then they go and seek out the additional million. So to you, you're just working with one insurance company via the retail broker in reality is you might be working with two or three different insurance or reinsurance companies to then support your your business insurance when we think about the facultative reinsurance market or the insurance market more generally you know you can segment that into large complex risk you can segment that into more transactional smaller smaller risks and you know essentially you know a hundred million dollar kind of insured value that's a small risk or a medium-sized risk.

2:48Dom Sindall: The large ones are more than half a billion going up to hundreds, potentially, for billions of dollars. Now, when a risk is large, time isn't necessarily a problem. Time doesn't matter so much. There's a lot of detailed analysis that goes on. There's a lot of negotiation. And it could take seven days. It could take 14 days. It could take longer to formulate a facultative reinsurance placement to support an insurance client of ours. Now, the smaller and the more transactional the risk, speed does matter. Price becomes less of a issue. There's less time to negotiate. And, you know, days doesn't necessarily work.

3:28Dom Sindall: It has to be hours. Otherwise, you know, the insurance company has missed opportunities if they're spending more than, you know, an hour or two on a particular submission, especially if it's small. Now, you know, the problem that we're solving for here is speed. If you can make facultative reinsurance frictionless and you can compress that time from days to hours if not minutes what opportunity does that open and how does how does the facultative market if it can achieve that play a role in that and what happens in the market if that type of financial vehicle becomes commonplace that's the problem we're solving for and that's what we'd like to talk to you about

4:06Stephen Sheehan: today well thank you because that was a very helpful explanation and i wish someone had given that explanation of facultative reinsurance 25 years ago. But just to sort of restate it, because I think what is important about this is that your client's client is a condominium holder, but your client is the insurance company. And then we talk about speed and then the friction. So what we're going to be talking about is how to provide the speed, basically, with all the sort of necessary analytics and things. But what's causing that friction these days? Why is it taking longer for someone to be able to get a quote from the reinsurance market?

4:37Dom Sindall: Matthew, it's an interesting question. I think when we think about how placement operates in the insurance industry, in the reinsurance industry for individual risks, it ultimately involves people, it involves brokers. and you're communicating via email, telephone, call, having in-person or virtual meetings. And your reliance on people being present and available is often causing placement to take days as opposed to hours. Now, if you think about how long it actually takes people to work on a particular placement, if you were looking at the stakeholders involved, the actual time taken is relatively short, but it's the total elapsed time that causes kind of the main challenge.

5:23Dom Sindall: So when we think about what needs to happen in order to create a frictionless environment in the facultative space in particular, it's, well, inevitably you do need to take the human out of the loop to a degree here. Now there needs to be controls, naturally, that need to be in place both on the broking side and the underwriting side. As soon as you put people in the loop, you're then kind of creating risk of friction to then okay whether that's going for one hour to four hours or four hours to a couple of days that's kind of primarily you know what's causing this now that's nothing new right this is this is a market that's existed for a long time and it's operated like that for a long time i think what's interesting about kind of where we're at is you know how technology is changing and how that technology is actually entering into our industry and and driving innovation in ways where thinking about taking humans out of the loop when it comes to placement and underwriting risk is now something that you and i matthew and steven we would go okay that that makes sense it's it's discussed it's talked a lot but you go back 10 15 years maybe longer that wasn't really a reality we could kind of operate in but it is today and i think that's that's the key and when we talk about um kind of innovation you'll be surprised this may surprise you but go back to 2020 there were three digital trading partners in the facultative space you go to 2026 so five six years later we've now got more than a dozen and that's a really powerful signal that's kind of happening into the market and this is something that's for me for steven this is really exciting a really exciting time to be in involved in the facultative reinsurance business because the market is changing.

7:10Stephen Sheehan: So both of you very well qualified to solve this problem. Stephen, you've worked for 12 years with leading reinsurance brokers. Dom, you had 11 years in broking and actually four years working at a major startup scale up. Dom, coming to you first and just sort of drawing on your experience both in broking and seeing what it's like to be working or building with a company is actually coming in this from the outside. What are the main lessons that you're bringing to what you've done before into what you're doing now one of the biggest things i've learned is in the context of

7:38Dom Sindall: innovation which kind of goes back to your question innovation happens at the conflict between different topics and when we think of kind of insurance and break that into different topics what's really interesting what's kind of happening and what i'm learning is if you bring data and you bring analytics and you bring technology broking and underwriting expertise kind of all together, along with the product mindset, which is kind of what I picked up in the scaling startup, you can do really interesting things. And the attention to detail to challenge the status quo and to kind of think about how you can reinvent workflows and how, in this context, business could flow through the market, opens your eyes to kind of new opportunities.

8:23Dom Sindall: And I'm quite open-minded in that sense and if i was to say anything like the insurance industry some argue could argue hasn't changed in in many many years but it is receptive to these conversations and it is now you know with evidence being receptive to change and it's it's really interesting and it kind of really keeps me going and gets me up and motivated every day i go to work steven so

8:46Stephen Sheehan: there's another question for you then in terms of what gets you motivated gets you up going for work in addition to your broking experience you cut your teeth in the insurance world or related insurance world in a catastrophe modeling company what are you sort of bringing from your past experience to what you're doing today?

8:59Matthew Grant: Matthew I could repeat what Dom said we've been fortunate enough to work very closely with Dom from the start of my career pretty much so had very similar experiences over the last 13 plus years I cut my teeth very much as an event response intern and paying attention to what was happening in the globe from a natural catastrophe perspective and writing blurbs and that was the transition for me from a natural science background into an industry background and since then I've been very fortunate that I've worked at two of the largest reinsurance brokers in the marketplace and I've been fortunate enough to be in a position where I've helped shape the data standards across both of these organisations because I'm within the broking team I've seen firsthand how the data flows through the placement process and into the reinsurance marketplace and how it's then used downstream for decision making that intersection point between analytics technology is a key area that really motivates me i very much enjoy waking up every day and tackling these problems well i'm delighted that the you're

10:23Stephen Sheehan: now seeing a revolution in the um in the data world because it seems like it's taken a long time to really get to the grip of these data issues which in theory you know something looking it externally might go well you know the client why isn't the data any good so are you now seeing a sort of meaningful shift now in the ability to actually get the data you need from your clients and then get it out to the markets that are providing the reinsurance to them when i first

10:47Matthew Grant: entered the industry 13 years ago i was surprised how document heavy the ecosystem was for insurance and the last couple of years especially we've seen a real acceleration in what ai can offer from an automation perspective. I'll pick two data types that we deal with. The first will be policy data. So policy data traditionally has been provided in the form of a PDF document or maybe a Word document or similar. These documents have been the way data has been exchanged from client through brokers into the marketplace and then into the reinsurance marketplace and downstream. They are ripe for automation and there has been a significant amount of investment in this space because they generally have a well-formed structure behind them.

11:47Matthew Grant: You are then able to leverage the ability of AI and other machine learning solutions to extract that data in a standardized way and then you can store it and use it for transmission further down the insurance value chain. So policy data we've made real leaps and bounds in that space in the past few years now if we look at another form of data so exposure data your information about the properties themselves that you're looking to get coverage for they're typically stored in something known as a schedule of values or an sov now these are typically an excel document but they can be other formats as well an sov that might have 30 tabs some of them in different currencies different languages and you're trying to combine that data together and again standardize it into a single form tech has grown substantially quickly where we're now in a position where companies can look to automate that moving towards delivering an automated solution for exposure data in a similar way to how policy data has been tackled in the past.

13:13Stephen Sheehan: So Stephen, just to clarify what you said on there, so the way you're solving this data issue is essentially using tools to make it faster and more efficient to get the information between the different parties. But you're not talking here about getting augmented information, the core of what you're talking about here is just making sure our data moves more smoothly.

13:33Matthew Grant: Correct that's right Matthew as brokers we have to represent our client when you get into areas like data enrichment etc that's when you need to be careful that the risk isn't misrepresented and for us in facultative we are very much focused on the transaction at least for the small and medium-sized risks and so the flow of data and the transfer of data is is the key aspects. I'm conflicted on this subject a little bit.

14:05Dom Sindall: If you can compress the time frame and you can get something done in minutes, there's a commercial opportunity. But if you take three hours, there isn't a commercial opportunity. What's the right answer to that? Because it's not perfection, but it needs to be good enough. And the question is, what's good enough? And what's good enough for the insurance company? What's good enough for the broker? And what's good enough for the reinsurance company in our context? And that's the question, you know, we debate a lot because there's a middle ground to be found here. And the opportunity is really interesting because it opens so many doors.

14:37Stephen Sheehan: Thanks, Dom. Before I come back on that point there, you mentioned TIV, which of course is totally insured value if anybody doesn't speak insurance abbreviations. And this kind of gets us into the question about the technology you're building or using. The more clarity and visibility and transparency you give the users, your clients, their clients, about the assumptions that were made and where that trade-off is, then people have, they don't have complete information, but at least they've got insight into the level of information they're using. But just on that topic of technology, it'd be good to understand a little bit about how do you make that build versus buy decision, how are you packaging this up as part of the application you're offering?

15:14Matthew Grant: If we're thinking, let's say, exposure data specifically, you need to think about kind of three key areas for exposure data, and that's how you're getting from the raw unstructured data into standardized data and those three processes are ingestion transform and quality ingestion is the data from a document that is normally in an excel file but can be other formats and identifying the data block so understanding what parts of the file contain information and data that you want to capture then it's understanding what that data is there are many different ways you can write what the construction of a building is or how much property damage exposure is on an SOV and sometimes they can be very clear and you have clear columns that literally say construction or say property damage.

16:22Matthew Grant: Other times they can be more nuanced or you may have multiple columns that all look to say something similar. So do you want all of that data or do you want to make a decision at this point to take some of it but not take all of it? So that's the ingestion process. Downstream applications typically require a much narrower breadth of information. So you have to structure that data in a standardized way it gets very difficult to start finding data standards that will work for everyone because if you don't have clear data standards then what's your tolerance going to be like in terms of data completeness and data accuracy and even when you have your data standards the next question is well okay I know what I want to do for these scenarios where I'm given a paragraph for construction I know how I want to interpret that kind of data the next question is how much of that do you want to be fully automated edge cases or any scenarios where you will want a human to review the problem we're trying to solve is speed to market so automate everything you get speed but what's a trade-off look like are there false positives that can go through that process where you're leveraging AI or you're leveraging some form of automation tech that's making decisions based on data but the data it's making a decision off is not really proper qualitative data or quantitative data so as part of this exercise you want to be thinking about whether you want humans to interact with this process and at what points that is so where are the validation gates in your automated process where you want a human in the loop do you want your data team your underwriting team your brokers whoever you are do you want them to be making a decision or do you want a third party to make that decision for you and depending on the organization and the team you'll have a range of answers there so really the prep work is the most important bit because when you've done that you can then start thinking does this solution exist in the wider tech sphere or is it something we need to build something or can we leverage a bit of tech here a bit of tech here and we do a build in between so you get a a joint solution that is a bit of both build and buy.

19:10Stephen

19:11Stephen Sheehan: And Don, we're covering a lot here, but what haven't I asked you about that we should also be talking about?

19:16Dom Sindall: The biggest question in my mind is what's changed? Why now? Why are we having this conversation? You know, we've talked about friction as a fact. We've talked about data transformation. And some of these topics have been around a while. Data transformation has been around a little bit. But what's really interesting, the faculty of the reinsurance market in particular, there is this recognition that the demand for FAC is artificially constrained by the way the market works and if you do reduce the friction inevitably you should be able to increase the transaction volume and grow the market kind of ultimately so the mindset of people in senior positions within reinsurance companies within brokers they're recognizing this the fact that technology is now quite pervasive you know not just in our personal lives but it's a topic that comes up every day in our professional world and you know this desire to want to get involved in digital trading or ai is is like i've never seen you know you go back 10 years you know the types of conversations it was quite inward focus and today you're talking about how can we build an agent between one organization and another or how can we create this frictionless experience around placement without kind of humans involved.

20:34Dom Sindall: And, you know, I would say, like, going back to Stephen's comment, some companies, going back 10 or so years, they invested a lot in data standards. They built the solid foundations to which, when these other topics came along, we weren't talking about AI in a meaningful sense 10 years ago, but we started to talk about it five years ago. And if you already had that foundation, it was a relatively easy leap to then go to these next topics. And if you look at who's providing digital trading capabilities, you know, those first movers, they were the ones who were really involved in this kind of early on and did invest in the data.

21:10Dom Sindall: And as a consequence, they also invested in and focused on changing organizational culture. And so there's been a transition period that a lot of people have gone through and it's very easy for them to build. And there's actually a resistance to buy because they have this belief they can do it themselves. But if you're this second cohort of reinsurer that's seeing what the first movers have done and the opportunity that they've managed to attract as a consequence of getting involved in that, to catch up, it's going to take a lot, especially if you're starting from a zero base. And so the propensity with that second cohort is to partner because it gets them up and running relatively quickly.

21:50Dom Sindall: You can get up and running in six months if you wanted to. But if you're starting from scratch, you can't do that.

21:55Stephen Sheehan: It's a really fascinating topic and it's just great to see as things come together and as you can build a solution that actually works. Before I let you go, I do have to ask you for a recommendation, each either or both a podcast or a book. So, Stephen, back to you.

22:09Matthew Grant: So, my book is maybe not fully work-related, but it's definitely helped from a professional perspective. And that's called Don't Believe Everything You Think. It's by Joseph Nguyen and mental health has taken much more of a centre stage in the professional world over the last decade. And those of us who are in more of an analytical role, overthinking comes quite naturally and going into detail. This book has been really useful for me because it teaches you how your brain takes a thought and turns it into thinking. and it helps teach you the process that your brain goes through in order to get there and how destructive thoughts can happen and with that understanding you then can start recognizing that in yourself it's been extremely extremely from my perspective to to help manage any thought processes so anyone who struggles with overthinking this book is a great read and i

23:24Stephen Sheehan: massively recommend it yeah well thanks for sharing that steven i'll be intrigued to take a look at

23:28Dom Sindall: that dom what about you so mine is the innovators dilemma by clayton christian i would say a must read if you're thinking about trying to challenge the status quo and the challenges of doing that in an established business context are very real. And the hurdles that you have to kind of overcome and think through very deeply are kind of assessed and researched at such a length in that book, mainly from other industries, but you can draw parallels over to kind of the insurance industry around how you seed opportunities and value propositions and then how you allow them to flourish through time.

24:06Stephen Sheehan: Yeah, well done, classic, which I actually do need to read because I haven't read it. So that's also one to add to my reading list. And then as we wrap up, a couple of quick ones. How should people get hold of you if you want to know more about what you're offering? And what's one thing you want to leave people remembering about we discussed today?

24:20Dom Sindall: So Matthew, for me, reach out to me on LinkedIn, Dom Sindel. I'm the only one on LinkedIn with that name. So it should be pretty easy to find. In the facultative reinsurance space, the first movers around digital trading have moved. And the second cohort of reinsurers are now emerging. so at Marshree we've built a platform that connects to all of these digital trading platforms that the the reinsurers have and we deliver one access point to our clients and that is about delivering frictionless fact to our clients so we're well on the journey it's it's been an incredible ride we've got a lot of work to do if anybody wants to learn more about this kind of going forward please contact your Marshree representative or if you have found me on LinkedIn, I'll happily kind of respond and we can set up a call.

25:08Stephen Sheehan: Great. Well, thanks to you both. Really enjoyed that. And I'll let you both get back to making reinsurance frictionless. Thanks.

25:14Dom Sindall: Thanks, Fati.

25:20Matthew Grant: Well, if you've made it this far, then I'm pretty sure you found that as interesting as I did. The Instech podcast comes out every Sunday morning, where we spotlight the latest news, Leading voices and freshest updates across insurance that you need to know about. If you would like to take part in these conversations, head to www.instech.co to find out how you can join our network and be a part of the insurance intelligence for the curious.