Name: Designing for insurability with FM: engineering resilience in the evolving data center ecosystem
Uploaded: 2026-04-16T15:16:12.905Z
Duration: 1 h 8 s
Description: Designing for insurability with FM: engineering resilience in the evolving data center ecosystem

Transcript for "Designing for insurability with FM: engineering resilience in the evolving data center ecosystem": Welcome, and thanks for joining us for the designing for insurability webinar. I'm your moderator, Lou Abramson, and I'll be guiding this conversation with Paul Heisel, VP and chief underwriter, FM Intellium, and Chris Long, business development executive. Paul, Chris, can you give us a little context on your background? Sure. Thanks, Lou. I've I've been with, FM for thirty six years, and as the chief underwriter for Intellium, I'm responsible for the global underwriting of our business. And this ensures consistency in our application of terms, conditions, capacity, and pricing. Over to you, Chris. Yeah. Thanks, Paul. So I'm not quite at Paul's FM tenure. I'm, coming up on ten years now. Today, I need business development for FM in the, DC, Maryland, and Virginia area sitting in our Reston office. Of course, we work with clients in a wide range of industries, but certainly being in the Northern Virginia area, a strong concentration of data centers as well. So to link my role to Paul's, I help identify new potential clients for FM Intellium and then work with Paul to provide them the coverage capacity and other resources that they need. K. So we all know data ecosystems today are being built faster, running hotter, and carrying more capacity. So who is FM? Where do we fit into this space? FM's approach to property insurance is powered by our partnerships and a mutual commitment to loss prevention. Through science based research and engineering, we drive change and deliver results that benefit our clients and partners and, in turn, their communities. Because as density increases and workloads reshape power and cooling demand, uptime isn't just something you manage once you're live. It's shaped by decisions made at day zero, and that's the context for Intellium, FM's data driven engineering led approach to helping data center owners and operators understand risk across the full life cycle, from site selection and design through construction to day to day operations. Intellium combines research, analytics, and hands on engineering to bring more clarity and predictability as your operations scale. So over the next forty five minutes or so, Paul and Chris will share how a life cycle approach helps your team keep pace with speed, scale, and density to avoid the kinds of blind spots that can surface after operations are already underway. So, Paul, let's kick things off with a question for you. What does an operator need to think about while still in the greenfield stage? We like to break the complexity into what we call the four qualities of a well protected data ecosystem. Not because data centers are simple, but because complex risk needs structure. So these four qualities are strategic design, site selection and construction, integrated protection, that's gonna be your sprinkler systems, operational readiness, making sure that you're ready to go, and skill teams and structured procedures so things run smoothly when you're operating. The reason this matters so early is that once you pour that concrete and lock in the layout, a large portion of your long term risk profile is set, you know, for better or for worse. So how do you see these qualities connecting to day to day operations? Insurability is engineered from the very first decision that's made. Geography, natural hazard exposure, site layout, and materials don't just affect the first cost. They determine whether long term risks can actually be controlled. When we look at, industry loss data, many losses trace back to weather exposure and a significant amount of outages involve human error. You know, the materials and systems you choose today dictate how the building handles extreme weather, fire spread, battery failures, power interruptions, even routine maintenance mistakes over the life of the asset. You know, if you account for those second and third order effects early, many downstream losses never have a chance to develop. Stable operations matter, because uptime think of it, it's revenue. But what's less obvious to that is a predictable risk profile leads to better insurance terms and lower long term cost of risk. That's a direct financial benefit. Yeah. And and, Paul, I would just add that from my perspective, it's not just about avoiding the bad outcomes. It's also about the operational predictability. So when you have engineering facilities and your risk teams aligned early, the day to day operations just naturally become more disciplined almost by default. So you're gonna have fewer manual workarounds, clearer maintenance playbooks, fewer surprises as well. And then, ultimately, that operational consistency is what insurers like ourselves, lenders, and rating agencies ultimately react to because it signals to them that a facility that's built to perform reliably under stress, not just under those ideal conditions. So as you talk about, you know, abilities to perform under stress, what what should operators expect from this approach in the sense of just real tangible outcomes? Yeah. What operators and tenants really want first is resilience. You know, stable uptime, control disruption, and confidence in the day to day operations. When you design for that level of resilience, the asset naturally becomes more attractive from an insurance standpoint. So the order, it it matters. Resilience comes first because that's what the clients and tenants value. That same resilience is what makes the location or business an attractive and insurable asset. We have, clients at FM who go the extra mile. They design and build their roofs, for example, to withstand higher wind speeds. Or they compartmentalize their data halls so not not all the GPUs are exposed to one catastrophic event. Chris, is there anything that you'd like to add? Yeah. Well, so so two things, and I'll I'll pack this from two different perspectives. So first, naturally from the insurance side, efforts like this show up very clearly in a number of ways. Right? So fewer claims, much narrower, loss scenarios, and then more predictable outcomes. And so that's ultimately what translates into better capacity, more stable pricing, and then fewer surprises come renewal. And, of course, I can speak to that very directly and through experience in my business development role at FM. And then second, from the operator lens, from the lens of the folks we're working with and and speaking to today, the day to day win of all this is simple. You're gonna have fewer cascading failures, fewer all hands events. When the resilience is engineered in into your facility, that facility is gonna behave in the way that your runbooks assume it will and gonna have a lot less deviations from that plan. So as we take a step back, Chris, can you help us describe the evolution of the data ecosystem? Sure. Yeah. So I'm gonna add a little personal flavor here. I've had a front row seat to the data center evolution for the past thirty years. In '96, my family moved to what was, at the time, pretty rural, Loudoun County right before AOL or America Online, moved in and the data center boom really began to a certain degree. So fast forwarding to today, we're seeing an unprecedented digital load driven largely by AI, by the Internet of Things, and, of course, streaming demand as well. And that's putting a constant significant pressure on uptime. So these modern hyperscale environments, they're denser, they're more interconnected, and they're built much faster, often with constrained, and tight supply chains. So that's creating a very different risk profile, especially as the design of these buildings shift from the traditional concrete bunker that we saw for data centers for years into these larger, lighter span structures. So, you know, tying that back to my own experience, the view from my windshield has changed dramatically over thirty years from, horses and farmland to the very, creative and, efficiently designed data centers that we see today that are sprinkled across Loudoun County and, of course, well beyond that. So as things evolve, what are some of the challenges you're seeing from the client's perspective in build and operation? Yeah. So certainly, the number of items there, but I'll try and tackle just, just a couple. So one of the biggest challenges we see starts early on in the build. The investors are under pressure to move fast, and we have speed driving these decisions. In many instances, unfortunately, resilience ultimately becomes secondary. So, yeah, faster construction, lower cost material that can look attractive upfront, especially from p and l perspective. But on the back end, they can lead to higher insurance costs, higher utility bills, and more operational risk. Ultimately, more than wiping out those initial savings, you might see at the beginning of this process. And then, of course, as, you know, Paul's already alluded to a little bit, and we'll get more into it. But once that facility is built, those risks are really locked in. And that's why early engagement is really critical here. The way we we tend to phrase it here at FM is, you're gonna need that engineering input before our project is is 30% complete or even earlier. That's really when you still have control over the long term performance and risk of that facility, because oftentimes, the cheapest decision on day one becomes the most expensive one over the lifetime of that asset. Yeah. Yeah. And to to add on to what Chris was saying, the ecosystem is carrying more load, more density, and more interconnection than it was ever designed for. When speed to market collides, it's a collision. So you have the speed to market is colliding with supply chain constraints. You're introducing a new risk profile, one where decisions made for speed can cause creep into long term exposure. Material availability and supply chain constraints are increasing the complexity of these builds and scheduling for that matter. So pre planning far in advance of the shovels going into the ground can pay dividends. Another point I would like to make is when your data center becomes operational and it's creating revenue, you know, you'll need a partner like FM to identify those gaps in risk that may have come from the considerations that I just talked about. So if we go a little further here for data ecosystem builder, what are some of the biggest risks that can impact operation and, of course, investment? Sure. So I'll focus on kind of the the frequency lens of this, Lou, and, you know, three three areas to highlight all the bit of a stack for each given my, former life as an accountant. So first is from an industry perspective and everything we've seen, power failure is the number one cause of downtime, the arch nemesis of our, priority, which is, of course, uptimes. Second, again, industry experience, extreme weather drives roughly a quarter of of all lost instances. So both your primary and secondary perils from a NAPTAP perspective, which we'll, we'll dive into a little further. And then, ultimately, the the people factor. Right? So human error from what we've seen accounts for about forty percent of all outages. So that's kinda the the frequency lens here, and I think, Paul will will jump into severity a bit. Yeah. So the the the frequency, it causes disruption, You know, these little nuisances that happen frequently. And when we're looking at FM's last fifteen years of loss history, your fire is the most severe in the data center occupancies and all these are insured by FM. You know, it accounts for over 40% of the loss costs. But let's think about what can result from a severe loss. You know, severity can lead to long recovery times and it could lead to, let's say, a reputational risk. You know, do you want to be in the news that your data center had a huge catastrophic loss and that's gonna take months and months to recover from? So it sounds like quite a lot of risk on on the horizon there with an industry moving incredibly fast. Chris, how can FM help the data ecosystem operators keep up with all of this? Yeah. So keeping up is a is a good choice word, or term. Excuse me. So that's the core down breaker, is the industry itself is essentially moving faster than the traditional decision making models we have in place we're really ever designed to support. So what FM what FM, excuse me, focuses on is helping data center operators ultimately to cut through the noise and focus on what's actually driving resilience over the lifetime of that asset. So what that boils down to is having a very clear engineered path, which means not losing valuable time debating or reacting after the fact, but rather you're aligning around the risks that matter most early and with very deliberate detection. Ultimately, that's where this life cycle, risk assessment becomes critical. It allows operators to see second and third order impacts, while there's still flexibility in that design and construction phase. Again, this goes back to the previous point of that early, early action, early intervention, from that design and construction stage. So that, ultimately, the speed we're looking for at the beginning does not turn into redesign retrofits and, and operational workloads, later in the game. But they ultimately, in practical terms, what FM tries to do is we're helping our clients to move fast without further compounding their risk. So we're embedding that resilience upfront so their facilities can scale, operate, and ensure predictability long term. So you talk about life cycle. Maybe it's time to to start digging into the process a little bit. Paul, if we were to start from a blank canvas, what are some of the elements to consider? The first piece of the puzzle is the site. You know, where is this located? And also power. You know, whether it's behind the meter, from the grid, or accommodation. You know, that's key when we're underwriting these risks. Water availability matters. The natural hazard exposures, like I said, the where where is this site located? But we also need to consider physical separation between buildings. Where is the fuel for the diesel generators located? And where are the UPS, you know, the lithium ion battery UPS systems? Where are those located? All that matters. Now if you do choose a hazard exposed location, you know, key, key thing to remember is awareness. It's and we can engineer for it. And let's talk a little bit about, cascading failures, things that can happen. So in Texas, and I experienced this myself, you know, historic freezes, you know, caught well run operations off guard, not because the risk was unknown. Everyone knows that it freezes. But because the systems, they really weren't engineered into the design. So during the Texas freeze, many data centers, you know, they just switched to generator power instead of waiting for the power to go off, you know, and that's good that's good, continuity management. You know, and there are many resources that we have at FM to understand natural hazards, like maps for wind, for freeze, hail, earthquake, and flood. Another critical element, I'd like to mention is maintainability. You know, systems need to be accessible, serviceable, and replaceable without disrupting operations. You know, that's resilience for the long haul. So as you're talking about designing for maintainability, how do you think about designing for growth? Right? As the evolution of technology happens, maybe there's a shift from cloud racks to gen AI racks or the inevitable shift to even greater density. Yeah. That's a good that's a good question. The the goal I would say the goal isn't to future proof for a specific technology, but it's to design the infrastructure so it can adapt without introducing new severity when change happens. You'll think about having a robust solid design for infrastructure that is interchangeable and adaptable for changes or improvements in these deployments that we see. So it sounds like a lot of these risks are issues we've always had, and that has construction material. Chris, why is the impact so much more relevant today? Yeah. So, well, you're absolutely correct. Right? These risks by nature are not new, in the insurance world. They're not even new to the data center space. What's changed, of course, is the scale and the concentration of that. Right? You see it every day on the news, on the Internet, in the headlines, the $5.10, $1,520,000,000,000 dollar data center campuses that are being announced. So, ultimately, in today's environment, we're putting truly unprecedented amounts of capital compute and this continuity into single sites and sometimes even to a single data hall. Right? So you have that extreme concentration of value. At the same time, capacities are enormous, higher power densities, more interconnected systems with less margin for error. And you combine that with a level of concentration with this modern speed to market pressure that we keep alluding to, and, ultimately, the impact of what may feel like relatively small design decisions have exponential, impacts down the line. So these choices we make around materials, layouts, hazardous exposures, that might have been tolerable in the past and wouldn't necessarily have a material financial impact for a company. Today, all of that is amplified 10 x or frankly whatever exponential factor you wanna use. And that's why, ultimately, these issues are so much more relevant in today's environment than they were five, ten, twenty years ago. So as we're talking about emerging risks and this concentration of value, I start to think about something that is is kinda what I find one of the more fascinating technologies on the horizon and that's SMR. Is this something that FM is starting to look at? Lou, certainly it is. We're we're seeing increasing interest in technologies like small modular reactors as demand for reliable low carbon power as it grows. From an FM perspective, I would say the key question isn't whether the technology is innovative. It's how it behaves under stress and what that means for severity, recovery, and long term resilience. You know, as new energy systems enter the data ecosystem, engineering discipline and early risk evaluation become even more critical. And this is a place where we're doing, lots of research. So from an operator's perspective, what tools and early decisions help the team feel confident they can keep a stable environment? Confidence comes from early detection and what I like to call graceful failure, not heroics. Very early warning fire detection lets you see small rain conditions before there's smoke. Off gassing sensing, it can identify these lithium ion battery UPS systems which can be very large. You can identify these issues before ignition. You know, what happens when you have ignition? You can get thermal runaway and that can be a really bad day. The data center infrastructure management, it gives you visibility into emerging system failures while there's still time to act. You know, so these systems, they just don't prevent losses. They buy you time. And time is everything during an incident. So, Chris, do you wanna talk about a little bit about power failure mitigation? So, pivoting a little bit to one of our, favorite topics in, in data centers, which is power. Right? So when we talk about resilience, power is really the starting point, because without it, nothing else is gonna happen. And then in the industry reports, we have this phrase that one is not. And so in a modern data center, basically, assuming that a single layer of protection will work exactly as intended every single time is is really no longer realistic. The systems today are just incredibly complex, and everyone on today's call, I'm sure, has seen that firsthand. Right? So you have utility fees, substations, generators, UPS, switchgear, controls. You know? The list goes on. And every handoff that happens during this process is a potential failure point. Right? So if you don't have the redundancy, if you just have that one handoff point, you're gonna fail. That's why, you know, one is not you don't really have that level of protection you think you do. And so the challenge isn't that redundancy doesn't exist. It's that the redundancy isn't always independent. So we see these single points of failure created by shared pathways, common controls, maintenance activities, or simply the assumptions that humans are going to respond appropriately 100% of the time when under pressure. So this one is none is is really a mindset, has to be very intentional, And you have to ask the difficult questions of, you know, what happens if this component fails at the worst possible moment. Right? That that black swan type event, whether it's during storm, during a maintenance period, or even during peak load, right, that absolute worst time for it to happen. Do you have the redundancies and the backup plans in place to mitigate that risk? And it's not just ultimately about adding more equipment and and adding more capacity. It's about the design, having physical and logical separation, and then disciplined operations so that no single mistake, no single failure is gonna bring your facility down and threaten that all important uptime that you're so focused on. So sounds like some of the other causes of downtime can come from emerging risks. Paul, if we're not protecting against these emerging risks, what types of problems could happen? Yeah. So one of the emerging risks that we have is these large lithium ion battery UPS systems and and they can fail. And it it really looks dramatic at first and that's the danger. If you have an like an older UPS system and, you know, it starts off gassing, you know, or you get some water intrusion or something, you know, this can all lead to thermal runaway. And again, like I said, it's a bad day. And then when you have this incident happen, you could have some, jurisdictional shutdowns, months of restoration and reconfiguration. And ultimately, like I talked about a little bit before, is this reputational damage. And reputational damage is not insurable. And if you ruin your reputation because you can't deliver on your promise to your tenants, you may lose that tenant. So my message is that the loss is not only that fire, you know, it's what happens after. So as we look at operational readiness, what can help backstop some of that downtime for operators? It's really building for resilience and maintaining it. Now as an industry, data centers are generally good at this. Critical systems we're looking at an n plus one integrating billing management systems with the HVAC. Also making sure that cooling capacity is built for AI workloads. Chris, would you like to add anything? Yeah. Absolutely. So, you know, one of the phrases that we, that we use often, right, is prevention is faster than reaction. And we've learned that ultimately, especially when it comes to data centers, proper prevention in the long run ultimately does outperform reaction, and this starts with discipline. Right? So most outages begin with that human element, not a catastrophic failure, which is why having clear runbooks, training, and change control matters so much in, in the data center space, especially. The next factor is a little more physical. Right? It's about having critical spares. So when those failures do happen, that we have the right sites excuse me, right components on-site to have that immediate remedy. And that's ultimately, a lot of times, going to be the difference between whether a rather minor temporary event becomes a larger, longer one. And so beyond critical spares, really, the last piece is having your facilities that are actually designed for maintainability. And what that means is you have systems that can be isolated, bypassed, and repaired without downtime across the board. So when you have this combination of strong discipline, the proper readiness for spares, and then a maintainable design, when this comes together, many failures never escalate to that point that you even know that anything happened. And so that's what we mean when we say these preventative steps ultimately are faster than reaction even if they take a little time upfront. So to keep us moving through this life cycle, as operators prepare to go online with a build out, what are some of the challenges and gaps that can come into play as we phase from build to operation? Yes. There could be gaps in the insurance. You know, construction and operations have different insurance needs. If for example, the gap can exist when the transition occurs in a building. So imagine a building is being built. You have data hall that is finished and now it's up and running. It's operational, but the halls around it are still under construction. So this phasing approach has both construction and operational risks in the same space. So understanding the demarcation where the construction ends and operation begins is key. Then when you have this situation, you know, one underwriter like FM providing both construction and operational capacity, you know, this can alleviate that stress. So as we enter the operational phase and and and the build team moves and and now we're we're day to day, how can uptime be protected in that new day to day environment? The I would say that the design that sets the ceiling. So that your design sets here's the bar. It's really high. But the people can determine the floor. So you could have the best facility ever, but if you don't have the right people, you could have some some issues. Training procedures and change management management, this all matters because about 40% of outages involve human error. You know, preventative maintenance, dedicated risk management support, and growth stage organizational models or controls, it's not overhead. Now a strong partner like FM can help fill knowledge gaps and reduce that strain by providing engineering insight and life cycle risk assessment that these lean teams simply can't generate alone. So, Chris, you mentioned you have a a background in business risk consulting. Maybe we can lean into that just a little bit here? Yep. So I promise it will be the last chapter of my life story. I seem to be, telling today between my move, my accounting career, and now business risk consulting. But as Lou set up, so far moving into the the sales role I have today, I worked on our business risk consulting group at FM, and one of the things we did was help our clients to develop, document, and ultimately test really strong business continuity plan. And, really, no industry, for no industry is this more critical and, frankly, more foundational than it is for data centers. Right? Having strong BCP ensuring that uptime is essentially a prerequisite for being successful in this business. So a resilient operation ultimately does start with having strong business continuity management system That includes, of course, having the right recovery strat strategy. Excuse me. Whether that's cold, hot, or mirrored sites. Right? That's a decision and an analysis that that each company needs to make on their own. Beyond that, having clear emergency response plans and not only having them, but having them for specific scenarios. Right? So having those plans in place for natural hazards, for equipment failures, for utility disruptions. You know, you need to plan for each of these somewhat, predictable scenarios and know what your people and your processes are gonna be doing, in response. And then ultimately, also a level of having very disciplined monitoring and testing of your, of your critical facilities. And ultimately embedded in all this is the human element that we keep coming back to, because as we've shown on the screen, right, your people are both your greatest solution, but also if not properly supported, they can become your greatest source of risk. Because the same teams that are keeping your facilities running are also the ones making these key decisions sometimes under pressure. So on that tone, just a a quick note for everybody enjoying our our webinar today. There are some additional resources in the doc section on the right of your screen. Chris, can you highlight a couple of those resources that people might wanna start with so they can stay ahead on best practices? Yeah. Sure. The great thing about, many of our resources that they because they are, excuse me, publicly available. So we, of course, have, the FM resource catalog that I would highly encourage you to check out. All of our FM data sheets are also publicly available. So that's when you can get really into the fun engineering details of exactly what is the right fire protection systems, for my facility and and way beyond, that topic alone. You know, touching briefly on FM approvals, which is another division of our organization, we have 28,000 FM approved products specifically for data centers. So when we talk about putting the right equipment, putting the right infrastructure in your facilities, that's, again, a decision, and a choice that we can help guide you on. So long list of resources that are available. I encourage you to check them out, whenever you're able to. Thanks. So we've sort of run through what the what the potential problems are and and some steps. What does this mean in practical terms for both predictability and long term stability? When when you put these qualities into operation from the start, you know, the strategic design, that determines your base risk, so that's your foundation. The integrated protection determines your reliability. Operational readiness, this determines your stability. Then lastly, your skilled teams, they determine whether resilience is sustained day to day. So together, they form the four qualities of a well protected data ecosystem. When those qualities all work together, you know, uptime becomes predictable and predictable risk is what every stakeholder wants. And what we what we wanted as well is another boring day at the data center. Yeah. So just just building keep using the same term that, that Paul does, but it but it's key here in the insurance world, in the data world, and that is predictability. That's what this long term stability is built on. And taking that life cycle approach ultimately reduces surprises in a number of areas. Right? So your underwriting process is smoother, your costs are more predictable, and these late stage retrofits can largely be avoided. And, again, every stakeholder is going to to benefit from this, financiers, tenants, regulators. And that only compounds further as the density and the complexity of these facilities increases, which, of course, we're already seen, and there's no reason to think that that's going to do anything but accelerate further. So, ultimately, when you have design, protection, operations, and people aligned early on, the risk becomes more repeatable and the asset stays stable over the long term. So I wanna thank you both for walking us through the life cycle today. For operators just starting out or continuing to think about how these ideas apply to their own sites or portfolios, where's a good place for them to continue the conversation and explore this approach in more detail? Yeah. So I think we have a couple of them already with some of the resources that we referenced, but more specifically to, data centers and linking this to, of course, our Intellium operation that that Paul helps, lead, I would point the audience towards, the URL you're seeing on the screen, which is fm.com/intellium. That's a great, starting point to access many of these resources information. And, I'll put myself out there as a direct contact as well. Feel free to, reach out to me as well. So we've had a few questions come in. If if you guys both have a little bit of extra time, do do you mind answering a few questions? Sure. Sure. Just for you, though. Thanks. So how do I make the business case to leadership for the ROI on improvements at this level? Yeah. That that's a a good question, Lou. And, you know, ROI, you know, the case it's not about eliminating loss. It's about avoiding this volatility that we don't want. When you can improve your site selection, protection, and operation operations, it reduces, you know, unplanned downtime. But loss severity when loss doesn't occur, it is not as severe. And also, when you make this investment, you know, there's less insurance friction. You know, you might get better limits. You got better deductibles. You might not have as many exclusions. And in long term, you know, this investment helps with operations, but also you don't need to retrofit anything. So you're really just trading a small upfront cost for a long term engineer proven predictability. So here's a question that I'm curious about. What does the data ecosystem look like in twenty plus years. Right? As as technology is moving incrementally faster and faster, where are we going? Yeah. That twenty years, you know, who knows what that future is really gonna look like. But what we can expect is greater density, tighter interconnection, and and less tolerance for disruption. And what that means is you'll have higher consequences. You know, the stakes are gonna be really high, but the higher consequences from design mistakes and there's gonna be less margin for human error. But who knows? Maybe in the future, robots will be running the data centers. You know? And there'll be greater dependence on early detection and automation. In in that future, resilience, it's not gonna be optional. It's gonna be assumed. Assets that aren't engineered for this predictable performance will struggle to ensure finance and lease. The operators who win, you know, in the future, this data center of the future, they're gonna be designing today for how the ecosystem behaves under stress, not just how it performs on a good day. And so the the fundamentals, they they won't change. But the cost of this, getting it wrong, it's gonna be much much higher. Yeah. Well said. So if I think about data center twenty years from now, few predictions. One, Paul, you'll still be underwriting PREPAD intelligent. Right? So everything is still gonna flow through you. Paul's Paul will be here. I'll probably still be in Loudoun County watching whatever the future is unfold. And from what I'm, you know, reading and seeing online, you know, we'll have some data centers up in space, maybe underwater, so we can deal with some, you know, new exposures there. On a more serious note, I am gonna kick the can a little bit and highlight that, FM, we're a very research driven organization. We have an entire team dedicated to researching every topic under the sun and certainly data centers, not only today, but the future of the technology, where is the industry growing growing and going, excuse me, and what are the the trends that we should be not only watching and keeping up with now, but also expecting to merge in the years to come. That's the expertise, the 100% dedication of our FM research scientists and data scientists. So, I highly recommend everyone, tap into that, and we know those folks are getting our clients and, hopefully, future clients prepared for whatever, the future of the data center industry looks like in, five, ten, or even twenty years. So, spicier to say, our FM clients will be prepared for whatever comes. Does building a data center to FM standards increase the time to go live? No. Not not inherently. In many cases, it actually avoids delays. But let's be honest here, delays tend to happen when there's late discovery, when you have design gaps or protection shortfalls or jurisdictional or insurance objections close to commissioning. So when we do the engineering and, do it upfront, you know, ideally before that 30% design, you know, we can just avoid all these late redesigns, the need for retrofits, and all these, different approvals that could stall everything. So it's not about slowing the project down. It's more about avoiding surprises that stop it cold at the end. Okay. So how does this resilience connect to the financing? Yeah. It's it's a good question. It's not the most straightforward answer, but, it absolutely does. At the end of the day, you know, it it all comes down to the dollars to a certain degree. So, lenders and equity partners in this space in particular care about downside protection and cash flow stability. So if you can show these parties, these financial institutions, that you have a facility that is truly engineered for resilience, you can speak confidently and, in most cases, actually quantify that you will have fewer unpredictable interruptions. You will recover faster when something does go wrong. God forbid it does. And, ultimately, that's going to allow you to have insured, such as FM, to provide you the capacity and the terms that you are looking for, which also bleeds into your financing, and and your lender covenants. So by doing all of this from an engineering and resilience perspective, it's going to lead to underwriting confidence, not just from that insurance lens that we're largely focused on today, but also to the, financing angle as well, that Lou's bringing up. So, Paul, I don't know if anything else to add on the the financing lens, but I know it's a it's a hot topic, in this space. Yeah. I think the these assets, you know, with the predictable performance, They're easier to finance and refinance and hold through the market cycles. And, you know, resilience, you know, it reduces perceived risk and perceived risk directly affects that cost of capital. So if just to sum it up, you know, engineering discipline, you know, upstream, it makes it the asset easier to underwrite downstream but for everyone. So before I let you guys go, if for our audience, if there was one one piece of information as a takeaway, what's what's the one thing you really want people to to take away from this? Chris, let's start with you. Yeah. Oh, man. One thing overall. I I would say, yeah, we we we talk plenty about being be thinking ahead and making these design choices early on. So I don't even think I need to handle that further. What I would say, honestly, is everything is moving so quickly, and our data center clients and and other operators out there are doing everything they can to meet deadlines, meet capacity. And so what I think that leads to is just naturally very limited time to work with and speak with folks outside of your own walls. Right? And so what that leads to is, I think, naturally, some of our clients are not able to stay as up to date as they'd like to with these emerging technologies, with new strategies, and and, methods that other companies are taking to mitigate risk, while further assuring uptime. And so what I'm getting at there, make sure you are working with a good external partner, whether that be an insurance carrier, whether that be an engineering partner. Certainly, at FM, we we can provide both those lenses. But whether it's FM or another company, just make sure you're working with outside organizations that have their, ear to the pulse of the rest of the industry and can keep you up to date on on new strategies and and perhaps new risks that you're not aware of. And then again, that gets back to the whole idea of planning for and mitigating those things in advance. Thanks. And and, Paul, what's your what's your big takeaway? Well, my big takeaway would be that, you know, uptime promises in this industry, they they don't fail, because the technology was wrong. They fail because the facility wasn't engineered, you know, to Chris's point. They weren't engineered for how it behaves under stress. So this predictable performance, it has to be deliberate. Delivered in the design, delivered in early detection, and disciplined operations. And when you put all those in place, resilience, you know, it it stops being this thing we're aspiring to. It becomes this measurable, tangible thing. Okay. That's about all the time we have for questions. If your question didn't get answered, just make sure to leave it in the chat along with your contact info, and we will be getting back to you. I wanna thank you both for a really cool conversation on on data ecosystem life cycle. And I wanna thank the audience for joining us, and, you know, no matter where you happen to be on your, resilience journey, take a look at some of the resources that we have, and, we hope to work with you guys soon. Thanks a lot for joining us today. Thanks, everyone. Appreciate it.