15 mins read
28
.
02
.
2023

What Everyone’s Talking About in Insurance with Matthew Grant

by Juan de Castro, COO, Cytora

This a shortened version of Making Risk Flow podcast, episode: “What Everyone’s Talking About in Insurance with Matthew Grant”. In this episode, Juan is accompanied by Matthew Grant, CEO of InsTech, to discuss the three major topics that everyone in insurance is talking about right now: digitising insurance to remove friction, the impact of climate change, and actively reducing or mitigating risk. Plus, Juan and Matthew touch on the potential impact AI and ChatGPT might have on the insurance industry.

Listen to the full episode here

Juan de Castro: Hello. My name is Juan de Castro, and you’re listening to Making Risk Flow. Every episode, I sit down with my industry leading guests to demystify digital risk flows, share practical knowledge, and help you use them to unlock scalability in commercial insurance. It’s great to be joined by Matthew Grant today. Matthew is such a driving force in insurance. He brings over 30 years of experience in innovation and more specifically around the topic of data and insurance. He spent over 20 years at RMS and now he’s the CEO of InsTech and also an advisor to a number of startups. Matthew, welcome to the podcast. It’s a real pleasure to have you join me today. Do you want to start with a brief introduction to yourself and what are you focused on at the moment?

Matthew Grant: Yeah, Juan, thank you very much and it’s a real pleasure to be here. And I’ve got to confess, I always find it harder answering questions than asking them. So I hope you’re going to be gentle with me. So, just very briefly, you mentioned my career, 25 years in catastrophe modelling. Really interesting for lots of reasons, but I think most importantly, it was an understanding of how insurers are able to use data and analytics to make decisions, which is of course what we’re looking at today in a whole series of areas. And then back in 2016 when we launched Instech London, that was really at that time to help early stage companies talk about what they were doing. In those days, it was difficult for people to have a voice that was starting to come out in the press, but it was an opportunity for people to get together initially in London at evening events and then that turned into a full business. Today we’ve got 170 companies. We’re working with both insurers and the companies that are creating data and technology of all sizes. Actually from early stage companies would likely to be working with Cytora as well as large organisations people know, such as Mastercard, Google or S&P.

Juan de Castro: I’ve seen the evolution of InsTech. It’s been just fabulous, right? It started as a place to showcase what InsureTechs were doing and now it’s become a real hub where really large companies kind of collaborate with mostly technology scale ups. So well done and congratulations on that journey. So clearly you are one of the few people that are really at the centre of innovation in insurance in London, but also more globally. So you’ve got access to both what startups are doing, but also to what insurers are looking for. So let’s start there. What are the two, three themes that you hear most often now in the market.

Matthew Grant: So we think as startups, as being companies formed the last three or four years, I think there’s an important distinction between scale ups, which Cytora is definitely in that category. There’s a really interesting theme now, companies that have been going for 10 years start to got some real clients, real products, real proven market fit. I think it’s important just to acknowledge that because for a lot of insurance companies they found it can be quite difficult to work with really early stage companies. Whereas when you get to an organisation that’s been around for ten years they kind of know how to work with insurance companies. In terms of themes, clearly there’s a lot of things happening .There’s three that I thought would be helpful to dig into a little bit more.

So the first is what I call removing friction. So this is where underwriting organisations are looking for ways to help their underwriters just be more effective at underwriting, move them away from having to do manual entry and all the other things that keeps them away from doing the job they’re paid to do. That’s the first one. The second one that we’re all familiar with, but it plays out in different ways, is climate change. And that ranges from the events that we’ve all known about for many years, such as hurricanes that are getting more frequent and greater intensity through to the shifts we’re seeing in areas that previously never got as hot or never got as cold. And then there’s another dimension to that we can talk a bit about in terms of what the regulators are asking for. So climate in all its forms is the second one and then the third one that’s taking a bit of time to develop. But we’re seeing really encouraging signs now where insurers are collaborating with their business partners, the corporate clients they work with to actively reduce the risk and actively put in place processes, tools, analytics, so that what in the past could have resulted in the loss is now being prevented before that loss occurs to the benefit of everybody.

Juan de Castro: So you mentioned remove friction from the underwriting workflow and processors and enable underwriters to be more productive, climate change and ways to actively reduce and prevent risk. So let’s go one by one and hear a bit more in detail. What do you see in the market? So on the first one on removing friction, tell us a bit more about that one.

Matthew Grant: Yes, so everyone’s talking about AI, artificial intelligence. It’s been rather overhyped the word. I think that is actually much more relevant and useful to understand what’s happening under intelligent automation. So intelligent automation is where you bring the technology together with an intelligent human to do the job. And what this means is that the role of the underwriter can be focusing on making decisions about what risks they want to underwrite and they don’t have to spend time doing the basic tasks such as declining business coming in that doesn’t fit some criteria or all the data entry, or even going to look for the data itself. So if you take a look back, if you look at what’s really coming through to the fore in terms of organisations that we can see being successful, they’re not where you’ve got some remote AI tool that’s doing everything for everybody that really still just doesn’t work for lots of reasons. Although it has some fun talking about Chat GPT, it’s really about this idea that you can create the analytics and you can create the data. And as you all know from what Cytora does, that straight through processing that enables both removing the friction and also importantly, increasing the speed of the throughput for the underwriter.

Juan de Castro: I also see many of our clients talking along the same lines. I’ve done previous podcasts with Roman from At-Bay, also with Thierry from AXA in Europe. Two things they mentioned quite a lot is this concept of human on the loop versus human in the loop. Right? So human in the loop is when an underwriter is required to do every single step of the workflow. Human on the loop is where technology can do all the or most of the heavy lifting in terms of capturing data, analysing data. But ultimately the decision is at the end of the process with an underwriter. And this resonates much better, especially in the mid market and large commercial segments, right? Where you don’t want to remove the underwriter, but you want to place the underwriting capacity where it really matters.

Matthew Grant: Now, it’s a great distinction, but I hadn’t actually come across that before. Human on the loop versus human in the loop. It brings to mind for me one of these sushi restaurants where the sushi is moving in front of you on a kind of revolving conveyor belt and you sort of pick up the sushi you want and eat it. And I can sort of visualise now the underwriter being on the loop, picking out the sushi dishes they want to eat, ie. the underwriting risk as opposed to in the loop for having to actually make all the dishes and decide if they want to use the ingredients. That’s a very elegant description of the distinction

Juan de Castro: At the end that is the concept of flow, which is also the theme of this podcast, Making Risk Flow, is you have risks flow through the organisation. They are only picked up when you need expert judgement in different steps. Going back a year ago when I started the podcast, my first episode was about making an industrial revolution in insurance and it’s evolving from manual processes to this type of flow business. In your experience, how far in that journey do you see insurers? Is it still like a promise or do you see actual insurers deploying this type of approach?

Matthew Grant: Yeah, if you look at the shift over the last seven years, which is when the term insurtech was coined, the mindset has shifted a lot. So I would say you won’t come across a senior leader at an insurance company who doesn’t understand the importance of this and understand the importance that technology has got and importance data has got. But then your real question, which is how far has it got? It’s still a struggle. As we know, legacy systems make it difficult to integrate the data into the workflow. And if you can’t integrate the data and the analytics into workflow, basically your solution is actually creating a new problem. But having said that, there are some quite sophisticated tools out there now that are able to tap into legacy systems and extract the data, do the analytics, put it back into the legacy systems. So I think what we’re seeing is a sort of parallel approach here. One is actually the technology and the systems themselves are being developed from scratch. And then the other one is people are building the tools to wrap around the old systems and at the end of the day, from the point of view of the underwriter, they can come in the morning and do their job. And to some extent they don’t really care whether it’s an old system or new system. They just want to be able to do their job, identify the risks, they should be underwriting and make money for the business.

Juan de Castro: Going back to your point about AI. So I see those processes as just a workflow, right? AI will have a role to play in different steps of that workflow, but it’s not going to take over, at least for the time being. And then you also mentioned ChatGPT, right? So we are seeing, for example, some specific use cases which are a good use of that technology. So for example, in our platform we can provide a summary of the broker email to the underwrite in just a few sentences, but those are specific use cases or steps in the workflow rather than taking over the full underwriting.

Matthew Grant: I was joking about this. Somebody, if you try and go in to use ChatGPT towards the end of the day in the UK, you get a response saying, sorry, we’re busy, please come back later. So I think it’s already discovered the bad habits of some financial organisations like I’m choosing not to answer a telephone call! But I think your point, I mean, in theory you could have a fully AI enabled underwriting operation and to some extent there are tools out there with things like Ki doing that. But the reality is, unless the data itself is completely reliable and integrated into your clients business, then you’re going to get big gaps and big holes in what you’re trying to analyse. I would say what is the barrier to moving into a totally automated underwriting operation for all but the most simplistic risks? Or I guess what we’re used to when we buy our home insurance. It’s that comprehensive way to be able to collect data that’s actually quite difficult to get in a commercial setting. Buildings are complicated, activities are complicated. So I think certainly in my lifetime you gave away where I am in my career, my 30 years of previous existence. What I wouldn’t worry about is at this stage of my life as an underwriter that I was going to be displaced by some technology.

Juan de Castro: The second one you mentioned in terms of themes was climate change. Let’s narrow down the whole climate change challenge to what are the things that really concern insurers.

Matthew Grant: It’s a big topic in reality, and people are quite casual in their word of the use climate. We tend to characterise in climate today, which most insurance decisions are made over a twelve month cycle. And so if you’re making decisions about which risks to underwrite or how to price, generally that’s going to be based on what’s changing in the climate today versus the future impacts of climate change when you get sea surface temperatures increasing, rising flood levels, all the things we’re hearing about with the representative concentration pathways that you use to measure that. So if we think for a minute about short term, we’re clearly seeing evidence of areas that are getting hotter and areas that are getting colder. So we’ve got things like the heat stress in Canada last year, the freeze in Texas. That’s one element where this is playing out. So areas that people hadn’t previously thought about are a problem. And then it’s not just the areas, but actually also the behaviours. So you can imagine if you live in an area of the world that you get snow for three or four months, a year, you’ve got a pretty good discipline for how you drain off your water if you’re not in your property and you protect your pipes. If you’re living in Texas, you probably don’t. So when Texas gets a freeze, they’re much more vulnerable to damage by the freeze. So not only is it a question of pricing, it’s actually much higher damage. So we’re seeing that. We’re seeing wildfire coming, we’re seeing more floods, we’re seeing more drought. And then on top of that, you’ve also got other effects coming in which is not directly related to climate change, but it is a factor of it, which is to do with litigation and some of the really huge costs of litigation going on in the US. And particularly in Florida now, where people are looking for others to blame or to increase the cost or not to pay for the claim.

Juan de Castro: You mentioned two implications there, pricing and litigation. So that’s the change in temperature and whether in those regions it’s a fact at this point. How are insurers thinking about analysing and pricing differently those risks? What data are they using?

Matthew Grant: Yeah, I think important things to understand with if you take in my background in catastrophe models for example that how does the catastrophe model work – does it reflect the future climate change? This model has been built over 30 years and certainly we are seeing climate change and so the hazard component, those models needs to be considered to reflect those changes. But there’s a lot of very good work being done that continues to be very valuable around vulnerability. So your vulnerability means if you’ve got a wind speed of a certain miles per hour then different buildings are going to perform in different ways, different fingers are going to blow off over 100 miles an hour, the roof might start blowing off. They’re still very valuable models. They need to be considered in the context of that. The frequency as I mentioned before, of the windstorm if we’re talking about hurricanes is going to change. And so the more sophisticated users of these types of tools are working with the providers, both the traditional providers like RMS or Verrisk or CoreLogic as well as some of these really interesting new companies coming in like Fathom. They’re working with them to understand what the models are telling them. And most important is theme that keeps coming back actually and it relates to so many areas of using analytics and data is understanding uncertainty and being aware of the uncertainty and being able to make decisions given uncertainty as opposed to the expectation that everything gives you a single number, and you use that without any kind of acknowledgment about how it may have been assessed or what assumptions were made coming up with that output.

Juan de Castro: We see these probably the maturity phases in any other lines of business recently with cyber where almost the first step is just understanding what’s happening, understanding what are the changes, then it’s embedding those into the models to understand okay, what are the implications and then you start pricing and monitoring constantly. Would it be preferred to say we’re on stage two?

Matthew Grant: Yeah, definitely adapting. If you look at the main models they’re now creating what’s called climate conditioned models. Give some I had experience back in the RMS past looking at decisions about how to model possible increases in future climate risk because of seasonal variation in hurricanes. And it’s very difficult even to the point that people arguably we shouldn’t be doing it, of trying to adjust the frequency of catastrophic event based on a perception of increased activity for hurricanes because there’s a whole range of different time periods of where we say hurricanes, you get that range of activity and frequency. So one year you might have no hurricanes, the next year you might have like we did this year, really intense hurricane coming in with hurricane towards the end of the season, next year maybe multiple hurricanes. So you get these short term trends move around a lot and then you get the long term trends pointing in a direction where yes, the overall losses is going to go up. But if you’re trying to make a decision as an underwriter year on year and trying to account for this seasonal variation, it’s not as difficult as actually impossible. And people can get into a lot of trouble by trying to follow that bouncing ball literally too closely. So that’s where it comes back to my point about, yes, the providers of tools analysis can provide some information, but it’s really, really important that you’re back to your human on the loop. The human on the loop in this case needs to be somebody that is actually doing what underwriters should be doing. They understand their client, they understand the hazards that can impact their client, they take the input from third parties. But ultimately, the reason they’re paid to do their job is because they can take all that information and use what you might describe as facts or certainty, but also use their own internal heuristics and work out what is the right thing to do for that particular client without being over reliant on some independent third party number that they haven’t got any senses to really where it came from.

Juan de Castro: But that’s also why it’s so important to be for insurers to have the capability of embedding new data into the risks they are saying. Right? Because as you said, the first time you see a new risk after hurricane, you cannot make a decision on a one off event. But what you want to do is make sure that you’re capturing data so that in the future you can start having more meaningful correlations. That is something we are also seeing many insurers doing now, which is almost in that first phase about can you capture the right data about the risks and the outcomes so then you can inform those judgement calls.

Matthew Grant: Yeah, and when I come back to that, I’m very glad you mentioned outcomes in that because I was interviewing Parul Kaul-Green, who’s Chief Digital Strategy Officer at Liberty and she talked exactly about that point, which is in her role, which is working with the underwriters to help them make better decisions. All things we’ve talked about, she’s really focused on outcomes. And her point is, don’t just think about output, meaning you can work really hard and get something out, but if that doesn’t have a business outcome, then you’ve been wasting your time. And I think that’s often what people that overlook with data and analytics is you’ve got to get data that actually allows you to make a decision. If you can’t make a decision on the data, you’ve kind of wasted your time. And I want to come back to your point there about multiple data sources because that’s also really important and I know it’s something you’re increasingly doing because I heard from Richard Hartley when we interviewed Richard as CEO for the podcast about the partners you’re working with, many of whom we know well. But yes, we’re definitely hearing from insurers. They want that flexibility about who they can work with and bring the data in. And I think some of the more forward thinking data providers are actually agnostic about who they provide data to. So take JBA on flood modelling for example. They’ll provide data directly to their clients, but they’re also very happy to embed it in a platform and give the insurers choice. And what it means is insurers can then decide. Sometimes they might want to have multiple data providers for flood. Sometimes they’ve got a preferred one. Sometimes they might want to drill into the data and if it’s like a really big underwriting decision or big uncertainty or a property that’s very close to river, for example, really dig in, understand what’s driving those numbers, doesn’t happen all the time. But they want the flexibility to be able to do that and they don’t want to have to go and swivel all their chair and go to a different system when they want to get their flood data versus the one they’re using for their underwriting workbench.

Juan de Castro: Definitely. So we talked about pricing and relationship with climate change. You also mentioned litigation. What did you mean by that?

Matthew Grant: There’s an extraordinary thing going on in the US now called litigation finance. And I only came across this recently when I was talking to Bob Reville, CEO of Praedicat, which is building casualty or liability modelling and what’s happening in the US. And there’s lots of drivers around this litigation, but this one in particular I just thought was extraordinary. People are actually creating funds to raise money to go and sue or to go and contest claims and this money is being put that to challenge the insurers, to get insurers to pay more for a claim and bizarrely, is actually inadvertently being funded by the insurance companies themselves whose investment department are coming across these specification finance funds are run by very well respected groups. And the insurers on the one side of the business are actually funding them and then they get hit on the other side by these organisations that are deploying funds to go and pay lawyers to go and find out how they can get insurers to pay more for claims. So you get this massive escalation of claims cost and we’re seeing this from hurricanes as well. It’s not just limited into liability. And actually, one given your background, when you’re at Hiscox one of your colleagues at the time, Bevis Tetley, who was an underwriter, I remember him saying that one of the things that they were looking at when they were doing underwriting in Florida is they’re actually looking to see what the proximity was to a lawyer’s office, to the individual that they were ensuring, because they felt there’s a correlation between the proximity to a lawyer’s office and the likelihood for someone to litigate on the claim. I don’t know how true that is in practice. It’s not a new problem, but it seems to be getting much worse.

Juan de Castro: Fantastic. So we’ve talked about removing friction and this underwriter on the loop versus in the loop and then climate change. The third area you mentioned the very beginning was prevention. How can insurers help their clients in actively reduce or mitigate risk?

Matthew Grant: I mean, as many people know, this has been happening for a long time. Hartford Steam Boiler in the US now own by Munich Re and Factory Mutual. I mean, they started off over 100 years ago helping their clients, you know, the clues in the name stop their boilers from blowing up by better boiler maintenance. And yeah, that’s been ticking away in the background for sort of some of these more complicated, extreme risks. But what we’ve seen in the last ten years with the proliferation of IoT and data and sensors is insurers looking for ways to support their clients. And it happens both on personal lines and in large corporate space. Help their clients reduce the risk before it happens. It’s actually been a lot harder than everyone expected. We don’t need to go into the reason for that, other than the fact that ultimately there needs to be a benefit for an organisation, large corporation, to share data with their insurer and someone’s got to pay for either the sensors, although many of those were deployed, or the way that that data is aggregated. But what we are seeing is some pretty significant strategic initiatives that are bringing this right to the centre of how business operates. So the AXA announced back in June last year, what it calls its digital commercial platform. And this is actually supported by the CEO of AXA-XL, Scott Gunter. And what that is doing is essentially it’s bringing together some of the things we’ve talked about. So climate, but also AXA Smart services where they essentially have the goal of reducing the claims they pay to their clients. And that’s not through the sense of they’re going to get more aggressive about which claims they pay. They want to reduce those claims by providing these services to their clients. And they’re actually looking at making this as an area they can actually drive revenue from because ultimately, if they do this properly, they need to be able to generate revenue to support their clients. But at the end of the day, their clients get a service where they’re both getting access to tools and analytics to help reduce loss. And then the insurance costs can go down and the losses go down. So again, everyone benefits. So we’re starting to see a shift towards that. One of the ones that surprised me when I looked at the data was the amount of flood loss caused by escape of water. So escape of water, as you know, one is when your pipes break. So back to our Texas free example perhaps. Escape of Water flood losses in the US generally are about the same as flood losses from external rainfall or other kinds of flooding events. So pretty significant loss. You can prevent those by installing water shut off valves in the building but they’re very difficult to put in once the building has been built for various reasons. But if you add those in at the time the building is being built so basically what happens is they detect a lack of pressure in the water because there’s a leak, the water valve shuts off, stops anything getting any worse and it vents the flood before it gets too bad. Then that’s where you start to see a slightly slower development because it relies on adding these into new build properties. But if you come back in 20 years time, then the majority of properties are going to probably have water shut off valves because they’ve either been built or refurbished in that time. And so I think we’re seeing a combination of a greater awareness on both sides, insurers and their clients. But also technology started to find its way into the buildings and equipment that we use all the time and so you’ve actually rather to go and deploy expensive sensors and we all see this in mobile phones. You’ve got a sensor network, you can tap into it and then you can much more cost effectively do something with that to reduce the loss differently.

Juan de Castro: That is something we also looked at back at Hiscox and I think you nailed it when you said this is an obvious future. Right? It’s like it’s a clear win win. The client will have fewer or less severe claims, the insurance company will have to also have the loss ratio will be healthier. So it’s one of those areas where it’s a clear win for both sides. But what is preventing the industry from making progress in this space? And you mentioned clear benefit from the client. I think we can see that. You mentioned who pays for the devices. I get that and also the challenge around installing those valves in existing buildings and only being able to do it in new ones. But at least back from my experience at an insurer, one of the challenges was how do you use that data? Did you see that also as being a challenge to overcome for this to gain broader adoption?

Matthew Grant: I do. Although before that, I want to tell you a story because I was probably actually when you were in your days of Hiscox. I still am actually a cloud of Hiscox for my home insurance and we got free device to put onto our pipe to see if there’s a flood loss. And it was just because it was kind of complicated to install. I’m afraid that device is still sitting on the shelf in its box and hasn’t been installed, which just gets the problem. If you don’t build it in at the beginning, it’s very hard to deploy it. But back to your point about data. People kind of throw around this term, data is the new oil. And my best way of sort of answering your question is, data is a new oil in the sense that the black, sticky stuff that comes out of the ground is oil. You would want to stick it in your car. You got to do a lot of work to process it and refine it. And it’s kind of like the equivalent of turning that GUI stuff into something that works in the same way you take all that massive data and you turn it into something you can make a decision around. So, yes, absolutely. And I know talking again to Richard on the podcast. That was an area that he was talking about. As you’ve developed Cytora, it’s really interesting to see how the ten years you’ve been going, you’ve evolved into that space of recognizing that your role is to help companies understand all that data and make decisions out of it. And I think the second thing I’d add to that point is that the data that people trust the most, frankly, if they are doing a good job of collecting the data, a quick caveat is their own data. So if you’ve got insurance company or their clients data, so if you got insurance company and it can work with their clients, and you’ve got all the appropriate protections in place, then that is the most powerful data you can use to make a decision. I mean, it’s very useful to get third party data and satellite data, and that’s definitely got a role. But back to your point. Yes, that data is extremely valuable, but you’ve also got to build your digital equivalent of a refinery to go and extract the information. You want to run your underwriting system as opposed to running your car.

Juan de Castro: I really love these analogies. I think they bring to life the actual challenges happening in the market right now. And then you mentioned this device that Hiscox provides at least seven or eight years ago. We were experimenting because there was one device that actually we might be the one you had that did require installation, like professional installation. And then the challenge is not data. The challenge is will clients actually deploy those devices. Then we moved into a new one. I think it was called Leakbot at the time that you could just clip on your entry pipe, trying to remove that friction of installation. But I think the solution is not just the device. It’s also like, okay, once you’ve got all the data, do you need a new contact centre that is receiving the signals from that and then processing that and triggering calls to the client if something goes wrong? This is one of the areas where probably this is not a core capability of an insurer. I think we’ll see more startups in this space just helping insurers just manage the logistics of prevention.

Matthew Grant: That point about call centres is actually a very good one, because I know Previsico quite well who are using surface water flood early warnings for organisations of insurers and their corporate clients. And also they’re now starting to add sensors in for big retail outlets, putting sensors into the drainage and actually measuring when the water is about to overtop the drainage. Some really interesting case studies about how people have actually got alerts, clear the drain, and they’ve stopped the thing flooding. But one of the challenges they’re starting to see is exactly to your point there, is that this is technology, not insurance, but they’re now having to create a call centre because once you start deploying this technology, people want to phone you up if it’s not working, or they got questions about it. So you’re right, whether you’re an insurer or a technology provider, you need to have those. And then I did want to come back on your Leakbot. One, I’m afraid to confess, one that actually I had the leap bot, and I don’t mind a bit of DIY, and it’s still sat on the shelf. And I’ll tell you the reason why there, and I think this is one of the challenges in insurance generally is that until we’ve suffered a loss or we know someone that suffered a loss, we’re not all we told we have to do it. We’re not really motivated to go and take action. So intellectually, I can understand why putting a leak bot device clipping onto my pipe and wiring up to the WiFi or something like that made sense. It was just never the top priority. The thing I wanted to do that day, because I never had a flood loss. And so I bet you if I had a flood loss, I’d be sticking it on the day after, but it’s too late.

Juan de Castro: That also brings a good perspective, which is, at least in personal lines, the person who needs to install the device is a person like yourself or myself. And we are not professional buyers of insurance. Right. We’ve got other stuff to do. What I say the moment you go to more mid market and large commercial right, there will be a professional team managing insurance. I think that will have a bigger incentive, most likely to deploy this type of devices. So do you foresee a higher pickup rate of this type of technology in mid market and large commercial?

Matthew Grant: I do. And I think what’s happening is the traditional role of insurance by a risk manager, which is sat in the middle of the organisation, not a large budget, the most they could do was negotiate their rates for their broker. In some cases, not all cases. We’re seeing much more collaboration now between that role, the CFO coming in from a broader financial perspective and also working facilities management. So these companies are realising with the increasing use of captives, for example, that actually they can make better use of their insurance, if they actually start to manage their own risks. So we’re really intrigued by what’s happening with RIMS in the US or AIRMIC here in the UK, where they are actively encouraging their members to understand more about what you’ve talked about, understand the technology, understand how in your own organisation, if you’re a risk manager, you can collaborate with your colleagues. Some of these things actually are now quite cost effective, but to some extent, they’re still driven by what they have to insure. So I say we’re now in the kind of early adopter stage with some of those organisations that are using technology. It hasn’t quite gone mainstream yet into the corporate buying decisions.

Juan de Castro: I think everybody would agree we need to all as an industry to make more progress, but we’re probably all quite disappointed by the progress in the last ten years. This has been really fascinating, Matthew. We’ve touched on three extremely different topics, right? But probably each of those ones is worthwhile. Matthew, fascinating chat as always. Thank you so much for joining me.

Matthew Grant: Juan it has been a bit of pleasure, and thank you for giving me an opportunity to talk about some of the things I’m passionate about and really enjoy listening to your podcast. Congratulations. What you’ve done, I know how hard it is to do this, and you’ve got some really fascinating episodes there as well. So thank you.

Juan de Castro: Making risk flow is brought to you by Cytora. If you enjoyed this podcast, consider subscribing to Making Risk Flow in Apple Podcasts, Spotify or wherever you get your podcast, so you never miss an episode. To find out more about Cytora, visit cytora.com. Thanks for joining me. See you next time.

Subscribe to Making Risk Flow podcast:

Apple Podcasts & Spotify