Greetings, and welcome to the Guidewire Second Quarter Fiscal 2026 Financial Results Conference Call. As a reminder, this call is being recorded and will be posted on our Investor Relations page later today. I would now like to turn the call over to Alex Hughes, Vice President of Investor Relations. Thank you, Alex. You may begin.

Thank you, Grace. Hello, everyone. With me today is Mike Rosenbaum, Chief Executive Officer; Jeff Cooper, Chief Financial Officer; as well as John Mullen, President, who will be available for the Q&A portion of today's call. Complete disclosure of our results can be found in our press release issued today as well as in our related Form 8-K furnished to the SEC, both of which are available on the Investor Relations section of our website. Starting this quarter and moving forward, we have also posted a quarterly earnings deck on the IR section of our website. Today's call is being recorded, and a replay will be available following its conclusion. Statements today include forward-looking ones regarding our financial results, products, customer demand, operations, the impact of local, national and geopolitical events on our business and other matters. These statements are subject to risks, uncertainties, and assumptions are based on management's current expectations as of today and should not be relied upon as representing our views as of any subsequent date. Please refer to the press release and the risk factors and documents we file with the SEC, including our most recent annual report on Form 10-K and our prior and forthcoming quarterly reports on Form 10-Q filed and to be filed with the SEC for information on risks, uncertainties and assumptions that may cause actual results to differ materially from those set forth in such statements. We will also refer to certain non-GAAP financial measures to provide additional information to investors. All commentary on margins, profitability and expenses are on a non-GAAP basis, unless stated otherwise. A reconciliation of non-GAAP to GAAP measures is provided in our press release. Reconciliations and additional data are also posted at the end of the quarterly earnings deck on our IR website. And with that, I'll now turn the call over to Mike.

Good afternoon, and thanks, everyone, for joining us today. Q2 was another strong quarter with ARR growing 22%. We continue to see momentum and demand increasing and the results across the board this quarter reflect what we believe makes Guidewire a uniquely durable business. Before I go into the details, I want to take a step back and provide my perspective on the position Guidewire occupies in our industry, the role we play inside an insurance company and why that combination creates long-term durability even in periods of technology disruption and change. Guidewire is the stand-alone leader in delivering mission-critical core systems for the P&C insurance industry. We are now a SaaS company, but understanding what our solutions actually do inside an insurance company is essential to understanding our durability. Insurance is a highly regulated trust-based industry that evolves deliberately and depends on precision, resilience, compliance and accuracy at scale. Guidewire sits at the center of that environment as the operational backbone of the insurer, embedded across the core operating functions of underwriting, claims, finance and regulatory reporting. Our platform supports the complex financial and regulatory framework that underpins the industry, establishing reserves, tracking premiums collected and claims paid and enabling a highly regulated structure that spans hundreds of integrated systems, millions of insureds and trillions of dollars in transactions. At the transactional level, we serve as the system of record for risk when a policy is written, when a loss occurs, when a claim is filed and paid. Those commitments and outcomes are executed through Guidewire. And today, we don't simply provide that software. We operate it as a continuously improving secure, reliable and scalable cloud platform that strengthens over time. The complexity of replacing a core system in the insurance industry means deal cycles and implementation projects are…

Thanks, Mike. Q2 was another tremendous quarter. We surpassed the high end of all of our financial outlook targets, and we are raising our full year targets across the board. Given the market backdrop, we thought it would be helpful to give a few incremental one-time disclosures to help investors understand the durability of our model. First, ARR ended at $1.121 billion and grew 22% year-over-year or 21% on a constant currency basis. Additionally, fully ramped ARR ended Q2 at $1.42 billion and fully ramped ARR growth continues to outpace ARR growth. Our market experience has taught us that we can maximize customer alignment and lifetime value by negotiating ramped subscription fees over a multiyear period. We quantify the impact of these ramps in our metric fully ramped ARR, which only quantifies the first 5 years of a contract. We typically disclose this metric annually, but thought it would be helpful to remind investors of the power of this dynamic this quarter. Second, we continue to see customers lean into longer-duration contracts and larger commitments. This shows up in a number of metrics. For example, the average contract term over the last 12 months for new InsuranceSuite deals is over 6 years if you look at the weighted average duration weighted by fully ramped ARR. We have seen this metric increase over the last 18 months as larger customers push for longer contractual commitments. As a reminder, our standard contract duration for new cloud arrangements is 5 years. This dynamic is further evidenced by RPO growth. RPO finished the quarter at $3.5 billion, representing 63% year-over-year growth. We generally do not talk too much about RPO because we tend to focus on the powerful recurring elements of our model, such as ARR and fully ramped ARR. But in the current…

Our first question is going to come from Adam Hotchkiss at Goldman Sachs.

I guess to start, Mike, I appreciate all the clarity on the core continuing to accelerate, but it would be great to understand how you think about what Guidewire's position in the broader AI stack looks like over the medium term. We hear a lot about competition outside of the core from forward deployed engineer models and disruptors deploying LLMs on insured data. So just maybe clear up for folks Guidewire's strategy as it relates to owning AI versus enabling AI and then how that impacts your revenue opportunity.

Great question. I appreciate it. And I would definitely say that it would be quite a bold statement for us to say we're going to own AI in the insurance industry. What we're going to own in the insurance industry is core systems that I am very confident in. We see that momentum, and we see that insurance companies need to modernize. They need these core stacks to work effectively. There's plenty of insurance companies that need Guidewire to own the outcome with respect to AI capabilities. But running an open model where we see other companies that are going to use other components from other AI technologies in and with Guidewire, it's absolutely part of the medium-term outlook. And I think that this is really very, very important to understand. I have had numerous conversations with Tier 1 CTOs and CIOs in our customer base over the past couple of months. And every single one of them stress to me that they expect there to be a mix of how they deploy these solutions in their environments. At the smaller companies and at the smaller divisions, more of this will come from Guidewire; at the larger companies, some of it will come from Guidewire and some of it will come from partners. This is going to -- this is an incredible time in technology. And I absolutely want to stress that where we are one of one, I think, is in the perspective that we're going to be the most trusted, scalable, reliable core system that you can do anything you want with respect to AI and Guidewire. Now like how do we -- how does -- what you say, what parts of this do we want to do very well and maybe someday own, I'll give you a…

Okay. That's great, Mike. Really, really helpful. I wanted to then pivot to the core. I know we've talked about 25% or so of premium flowing through Guidewire today. And it feels like AI is may be moving customers into the cloud more quickly if your fully ramped ARR is accelerating off of the 22% in fiscal '25. So what's your updated view on the pace that premium moves into cloud and where Guidewire's penetration ultimately gets to over the medium term?

Thanks for the question. I would say it's definitely improving. And as you heard us talk about with respect to the results so far this year, the results in the quarter, the visibility that we see into the back half of the year, both for new business and expansions and specifically larger deals at large Tier 1 and Tier 2 insurance companies. This is just extremely positive for our business. that's what gives us the confidence to be able to update our outlook. How that relates exactly to the percentage points of global DWP that flow through Guidewire, it's very difficult for us to say or project that. I don't really run the business that way. We look at it more from a net new ARR and net new fully ramped ARR perspective and the specific workloads, the specific lines of business that exist at each of our customers in each of the geographies that we support. And then we look at it in the end of the year, and we report that out, obviously, kind of at a yearly basis, how we've done. But certainly, it's increasing. And certainly, we see demand increasing. And I think demand is increasing because of the potential that everyone sees in generative AI. They see what they can do. Like I think you guys have all heard me say this before. What's so startling, what's so special about this technology is every single person that wants to can see how powerful it is because we can all use it in our consumer lives. Like we can all touch it, we can feel it, we can ask it questions. And then you can just immediately see, "Oh, wow, I can use this in my company." But you can only use it in your company if you're running on a modernized core system. If you're running on a core system from Guidewire with APIs that you need, with the MCP servers you need, with the partnerships that you need, that's what really unlocks this, and that's what's driving the momentum in the business. That's what created the quarter that we saw. That's what's giving us the confidence to raise the guidance for the year.

Our next question comes from Ken Wong of Oppenheimer.

Fantastic. Very clear, very assertive statements on the AI front today, Mike. I think those were fantastic. I won't belabor the point too much since I'm sure my peers will. I wanted to maybe focus on new products. You mentioned good customer feedback on PricingCenter. You signed your first deal. Would love to get some early comments in terms of what you're seeing in those engagement, in those conversations. And then any update on whether or not there's some traction on the underwriting side?

Yes. Thank you very much for the question. So PricingCenter is super interesting because what we're seeing is people really leaning in and wanting to engage with us to talk about what's the vision and specifically, how is it going to be integrated into PolicyCenter. So for a Guidewire customer that's running PolicyCenter, there's just this obvious connection between the product model, the way that we define the product model and how that relates to what the actuaries are going to use to be able to create the products that they need, how it connects to our data platform and to be able to provide the data they need, to create the models they need, to stay current, to compete, to adjust to what's going on in the market. There's a lot of engagement there. This is a deal cycle that's kind of long, though, right? This is a thoroughly researched, thoroughly studied. Sometimes there's a POC associated with these deals. And so it's kind of more similar to our core sales process where, hopefully, as we said, we closed one deal that was like more than 10 years. Hopefully, those deals won't last 10 years. But it is something that's going to take us a little while to build. We were excited to get that first deal done, but we're also excited about the amount of pipeline and the amount of engagement that we're creating for PricingCenter and for us to start to participate in this segment of the market. It's very, very exciting. And then you asked about underwriting. Like on the underwriting side, we're still in the process of working with a small subset of customers that have expressed interest in really developing with them a solution that maps to what is really just honestly a very, very fast-evolving approach to agentic underwriting, let's call it, is what exactly does that need to do with respect to receiving submissions from brokers and how do we map that to risk appetites, and then how do we ultimately map that to PolicyCenter. Lots of excitement and engagement in the market around this. We're excited about the product. And I expect over the next couple of quarters to be able to start to get this into production with a couple of customers and learning fast and evolving from there.

Fantastic. Really appreciate the color. And then, Jeff, just a quick question on the true-up comment, I think you mentioned still seeing some tailwinds from true-up activity. I think we on the outside probably worried a little too much that as DWP normalizes, you really wouldn't see any of that activity anymore. Help us kind of walk through the mechanics of kind of how that continues to be a tailwind for the business.

Yes. Thanks, Ken. Yes, we did see healthier true-up activity than we initially expected going into the quarter. That was a little bit of a tailwind in Q2. I think as we think about the remainder of this year, it's generally aligned with how we've talked about this over the last few quarters. We saw a very healthy backdrop coming out of the kind of high inflationary period that is tempering a bit, but we continue to see this activity. And the way it works is customers have premium baselines in their contract. And it's always been part of our model that as customers grow, they pass those baselines and then we have the right to effect a true-up order. It's not uncommon for some customers to buy a bit more premium than they initially need. So it may take and in certain cases, a few years to see a true-up order after an initial purchase. But we see pretty regular volume of this. We have enough of this in our model now that we can be pretty precise in our predictions. And this year, we do still expect it to temper a little bit off of the highs that we experienced a few years ago, but saw a bit of a tailwind in Q2 and the back half of the year looks pretty much aligned with how we expected it.

Our next question comes from Rishi Jaluria from RBC.

All right. Wonderful. Maybe I want to first start by following up on kind of the earlier question around perceived competition from AI. We've obviously seen both OpenAI and Anthropic announced kind of deals with some of the leading insurers. But at least on first glance, it seems like it's very much complementary and maybe even potentially additive to what Guidewire core and even some of the add-ons are doing. So I want to maybe understand how are you thinking about your ability to partner and work with the large LLM vendors and ultimately just drive greater customer success within the insurance industry? And then I've got a follow-up.

Super, super question. We absolutely see this as additive and helpful for Guidewire overall and the acceleration in the company. We have always run a very open approach to our products and to our ecosystem. We've always invited multiple parties to the ecosystem because we cannot and do not imagine that we're going to do everything for every insurance company everywhere in the world. Now obviously, Anthropic and OpenAI have this access to this incredible technology that has obviously changed and will continue to change the world. But we don't imagine that the work that they're doing is targeted at the deep, deep specific complexities associated with operating a core system in the insurance industry. And we think that leveraging the capabilities that these tools provide these LLMs or even these like desktop applications that sit on top of their LLMs, they're going to be most beneficial when connected to well-structured insurance processes running on modern core systems from Guidewire. And so we're very, very open to working with these companies. We're very open to working with our customers who have partnered with these companies around solutions that connect them to Guidewire. And like I said -- in the script, I said it a second ago, we see this as net beneficial to Guidewire because what you're going to be able to do with the Guidewire core system that's deployed, your operations are modernized, your operations have these connection points that these systems need. This is going to allow these companies to accelerate. This is going to allow these companies to become more efficient. And so we don't see this as competitive. We see this as additive to the overall demand in the industry for what we can provide. I think Rishi, John wants to say something here.

Yes, I'll just add a quick point. There's -- tied to your question in all of that context is the fact that insurance carriers and leaders of insurance companies are under a tremendous amount of pressure to drive pace themselves. So the ability to differentiate in the market that they compete in and sustain differentiation is under a tremendous amount of pressure right now. So the ability to work more proximate with them, solve problems with them and increase pace of innovation on top of the service and also increase speed to value in the way that they get to that first cloud implementation and consume products and services that we deploy, and also, to Mike's point, have the open architecture where they can do things over the top of that at the pace that they want to and need to stay differentiated is really driving a conversation with these carriers and leaders in insurance companies that get us every day closer to them, and that's what I'm most excited about is continuing to drive that proximity.

All right. Really helpful. Maybe just a quick follow-up. As we think about your kind of own internal AI development, your own kind of ability to bring AI to your customers, recognize you're dealing with a highly regulated industry where it could take a while to get that meaningful adoption. But the question I'd like to ask is, as you think about -- a lot of the focus is on efficiency, but do you see an opportunity to maybe even drive better revenue outcome and ultimately better customer outcomes for the insurers, leveraging AI? And what would that look like with your current roadmap?

Sorry, I want to make sure I understand. You mean better revenue outcomes for our customers?

Well, specifically that the insurers can generate better revenue outcomes, right, whether it's being able to have better quotes or service more customers and ultimately, the end people being insured get net benefits as a result.

1,000%, yes, okay? The insurance industry is an incredibly complicated thing, right, if you zoom out. It is structurally been sort of hamstrung by the amount of unstructured information and data that needs to be managed in order to effectively and efficiently conduct the art of insurance. And large language models attack this directly. They address this directly. So you can underwrite more efficiently, which means that you can look at more risk. You can evaluate more risk more quickly. You can manage claims, the input of the submission of documents and the conversations that you have to have with all the multiple parties can be analyzed more effectively. And so it's like those two examples are sort of like tiny little bits of why the underwriting process is going to become more effective and the claims management process is going to become more effective. And I think ultimately, the insurance industry, the insurance machine is going to become more efficient, which is beneficial to insurance companies and to the broader society and our economies. Like the insurance industry with generative AI, and I think this is why everyone is so excited about focusing on these kinds of partnerships with these big insurance companies is there is a significant potential to improve its efficiency overall, which, like I said, it will be -- I don't want to say revenue, but I would just say like the efficiency of these companies is going to improve. And we're excited to be a part of that, driving that and making that possible along with a lot of other companies, along with Anthropic, along with OpenAI, like there's going to be a lot of people that are focused on helping the insurance industry do this. Like John said, our customers are excited about the potentials here because for so long, you're sort of limited to the technology capabilities at hand. And now you have this new tool that understands natural language and can be taught to do things like underwriting and claims. It's really significant. So basically, 1,000%, yes.

I'll just hit on the daisy chain of kind of product strategy because one part of your question was product strategy. So on top of the core operating system, if you think about the pressure points, our customers need pricing agility, therefore, PricingCenter. That's why we take that step. Product speed to market is the next thing in that daisy chain that drives competitive differentiation for them, therefore, advanced product designer. And broker efficiency and effectiveness is the thing that's probably up for the most amount of transformation and disruption and enablement given LLMs and the models available and therefore, UnderwritingCenter. So it ties very closely. The investments we're making in the product strategy ties very closely to those things that are driving differentiation for our customers that sit on top of the core processing environment. So the fact that the core processing environment has an opportunity to continue to gain market share by line of business specificity and geographic specificity because of the rate at which we can deploy products and the components that we're putting out over the top of it, I think, are really good proof points for our strategic resilience inside of our customers.

Next up is Joe Vruwink from Baird.

Great to hear about the urgency to modernize. I maybe wanted to ask about the pace around that modernization. And there's been a lot recently even COBOL got its time in the sun a few weeks ago on maybe AI tooling, making it easier to translate. I don't think necessarily the translation of COBOL is the challenging part, but I want to get your take on just modernization timelines more broadly and whether Guidewire has the ability to maybe accelerate time to value because of their AI usage.

Yes. I'll give you a quick take on this, and I think John is probably going to want to add some -- his perspective on it. Yes, we're definitely working hard to ensure that our teams that are working on these migrations, both from on-prem Guidewire to cloud, but also the modernization projects are more and more efficient. And we're starting to see the early results of this in the actual projects. There's like a whole litany of different steps that are involved in one of these programs, and many of them can be enhanced and potentially even completely automated with generative AI. And so reducing that time line, increasing the pace of that, therefore, reducing the cost of those programs also helps us make an argument about modernization now. This is definitely an exciting component of the story at Guidewire. I would caution, though, that there is a certain amount of, hey, this is running on legacy code and this is running on a system that we can't support anymore. So this like one-for-one translate into something that's more supportable. I think that's an okay step. But it really doesn't get to what is very often a major important part of the modernization, which is rethinking your business process, rethinking your products, rethinking your approach to doing business, which is often part of a modernization. And that's what you really need to engage with companies like Guidewire and our ecosystem of SIs to really help companies work through that and get to a system that's modern, but also an operation, a business workflow, a set of new standards that really kind of set the company up for their go-forward operating model. So it's more than just the conversion of the code, but it's really the modernization of all of the activities inside of an insurance company.

Yes, I'll add the -- if we think about where we were maybe 2 quarters ago, and you got to think -- we have to think about this as the investments that Guidewire is making in our professional services team and multiplied by the investments that the SIs are making in their teams. And if we go back 2 quarters, there was a lot of investigation, a lot of discovery, a lot of proofs of concept, a very wide funnel of activity. That is starting to narrow over the last 2 quarters. We're starting to see some green shoots of some really impressive kind of percentage reductions of time to value. And the next step for us is to really continue to increase the velocity of those proofs of concept and early test cases to be rolled out as standard operating procedures in these programs. But there's an important additional step, which is rationalizing that with the SIs because I think, certainly, I've been in conversation with all of our SI partners. And there's no world where we want to be competing tool-based in what it takes to drive speed to value on cloud. So we'll be doing some rationalization with them and making sure that the tools are consumable by the customer base.

That's great. And then, Jeff, one for you. I appreciate the midyear disclosure on fully ramped ARR. I'd have to imagine there's seasonality in that number, just given the deal volumes in 4Q creating some second-half weightedness. Can you maybe frame how much of a given year's net new fully ramped ARR happens in the first half versus the second half?

Yes. I think there's -- obviously, you guys understand our business. You know that our seasonality is 4Q weighted. 2Q historically is our second strongest quarter, and we saw a very strong 2Q for us, and that flowed through to some healthy additions on the fully ramped side. But you'll have to wait until Q4 to get full gratification on that question. So -- and we'll certainly talk about it in the fourth quarter call.

Our next question comes from Parker Lane at Stifel.

Jeff, I appreciate the disclosure on ARR retention rates and the commentary on how a few million dollar-plus churn events you've had in recent years. Looking at the remainder of this year and more importantly, maybe your midterm targets, what sort of assumptions do you make or cushion do you bake in around ARR churn? Do you anticipate that things remain relatively consistent with historical trends? Or are you accounting for some incremental conservatism there?

Yes. I appreciate the question. And given our business, this is an area of strength of ours. We -- the assumptions are as we go bottoms up in every single account and have really good visibility into any sort of potential downsell risk that exists in our accounts. And the team flags all of those throughout the year. Usually, when we start the year, we have a good read. And so we kind of do that. And we try to be pretty conservative and cast a wide net on kind of how we think about potential downsell events. And then we usually end up performing better than some of those -- that wide net that we initially cast. But this is not kind of a top-down model assumption exercise for us. This is a very bottoms-up, customer-by-customer, account-by-account exercise for us.

Got it. And one quick one on ProNavigator. I believe last quarter, you said you were expecting $4 million of ARR and $2 million of revenue, 9 deals in the quarter. How is that trending relative to those expectations that you outlined last quarter around?

Trending positive to those expectations. I mean I was not expecting 9 deals in the first quarter. So we're thrilled with that progress. And we can think about how we will disclose that moving forward. But you should think about it as right now trending ahead of expectations.

Our next question comes from Michael Turrin at Wells Fargo.

I wanted to spend some time on the commentary on duration increasing. It certainly seems positive in terms of willingness of customers to commit to Guidewire. Maybe just speak more to what's leading to that longer duration. Are you finding core replacements show up as a prerequisite for some of the kind of longer-term AI-focused initiatives insurers might be looking at? Or what drives that? And as a small second part, Jeff, you referenced the backdrop is why you're giving some of the incremental disclosures, which we definitely appreciate. Is that just the software market backdrop you're referencing because your results seem generally unfazed here. So maybe just help frame why the incremental disclosures for us as well.

So on the first question, yes, this is 100% just because of the software market backdrop. And we felt that in that backdrop, some of the durability elements of our business were being missed. And so we thought it was a good time to lean into some of these disclosures that provide a bit more durability. I think Mike will probably jump in here. But on the contract duration, we always engage -- have always engaged in longer-duration contracts. There was a period of time when we transitioned to ASC 606, where we actually forced shorter contracts on our customers. And as we move to the cloud, our standard has been 5 years. In the early part of the cloud, if you look at duration, it was a little bit lower than 5 years. We saw testing the waters, wanting to explore smaller deals and see how it goes. And now with the maturity of the platform, kind of where we are on this cloud transition side of things, we have seen that willingness to lean in and make longer commitments, that trend has increased. And then if you look at the largest customers, in particular, the ones that are making really big bets on Guidewire, often that impulse is to move even beyond our standard 5-year terms and pursue a longer engagement. And we've seen that activity kind of more recently over the last 18 months increase.

Nothing to add. I think you got it exactly right, Jeff.

Alex Sklar from Raymond James.

Mike or John, following up on Ken's question on PricingCenter and ProNav and some of the early success there. Can you just reframe how you expect the adoption curve to trend and sales cycles you've seen based on what you've seen to date? Were these particular deals in the pipeline prior to the acquisitions? And maybe, Jeff, how did those initial deals look like in terms of uplift on ARR?

So I'll hit the -- I'll go on the first part and then Jeff can pick up the second. If I think about the ProNavigator deals, the adoption curve in claims, we're seeing the pipeline that's accelerated there has really been as that team came into the fold. And I just should say, while I'm on this call, I couldn't be happier with how that team has joined. The culture fit is great. The energy is exceptional. But as we think about our ClaimCenter customers that are on cloud, the receptivity to have the right conversations and start laying down tracks for what that looks like is what's really driving the acceleration there. There are conversations in underwriting as it pertains to ProNavigator, but the acceleration is really coming in the claims space. The PricingCenter piece, Mike mentioned a little bit earlier, which has a lot to do with those that are PolicyCenter customers and the integration of PricingCenter into PolicyCenter is something that drives a tremendous amount of value and a tremendous amount of appetite right now for the conversations. There are a lot of proof points. It is a big decision. Every one of these customers has some variation of pricing and rating inside their environment, whether it's ours or somebody else's. And so really testing the waters on that and pushing through some proofs of concept is important. But those customers that are driving pricing -- that are driving policy admin solutions that sit on Guidewire are really very interested in proving these things out and looking at potentially large and long-term commitments. There is going to be a lot of work to do to make PricingCenter fit all regions, all lines of business. So that's going to be something that we look at a lot of investment in over the next quarters as we go forward.

Yes. And on the ARR side, we haven't spoken too much on this topic other than to think about PricingCenter as a pretty meaningful ASP product. It's a little bit of a longer sales cycle. These are big investments that customers will make in that product. So we expect that pipeline to kind of build and transact a little bit slower, but be more meaningful and impactful. On the ProNavigator side, those are smaller price points at this point in time. But at this point in time, that tool or that product is primarily looking at standard operating procedures of an insurer. And it is our expectation to evolve that into other content areas that would increase the value of that product over time. So I think the price points that we're seeing today are nice starting points and when we should expect to grow those over time.

Jeff, maybe just a quick follow-up on Joe's fully ramped ARR question for you. I appreciate some of the unknowns around seasonality given the larger Tier 1 customer base. But in the first half of this year, was there anything in the fully ramped result outsized contributor either in terms of steeper ramps or larger migrations that kind of is abnormal for a first half for you?

It was an abnormal first half for us just in the fact that we -- the volume that we saw, some of the large deal volume that we saw was very, very exciting. We hope to continue to build on that. So I wouldn't say there was anything unnatural, but we are continuing to see the momentum build. There are a number of -- in the first half deals that were longer than even the 5 years. And so there's even some backlog that is kind of off of that fully ramped ARR metric. And all of this is just kind of continued momentum that we're seeing in the business. Last year, signing Liberty Mutual was a big event for us. And so that creates a somewhat difficult compare. But as we look at the pipeline for the remainder of this year, we have a lot of really interesting activity out there. So it's always hard to predict exactly when those larger deals will come in, but we're thrilled with the pace and we're thrilled with the traction, and we're thrilled with the pipeline.

Our next question goes to Allan Verkhovski at BTIG.

Mike, given the speed and innovation of what's possible from a coding perspective with AI, you've gone through a lot of investments over the years. You talked about demand for deployed services. Where are you making changes or leaning in more as it relates to your product roadmap? And how are you further adjusting it, if at all, your expected developer count growth over, call it, a multiyear basis?

Great question. So we're in the process, as you could probably imagine, of rolling out agentic development tools, call it, a harness that works effectively for Guidewire developers. And I should say for the folks in our professional services organization, the folks in our SI ecosystem and all the customer developers, we fully expect that these agentic development tools will be leveraged by our devs and everybody that touches Guidewire from a software development perspective. And -- for sure, we see this increasing pace over time for what we can deliver. We're a little bit early days to that approach, but the anecdotal feedback from the sort of first movers and the people that have really put their hands on these tools and figured out how to use them effectively is extremely positive and gives me a lot of confidence that the development velocity at Guidewire over time will increase, right? So then that brings up logical questions that like I had that you are asking me right now, which is, okay, well, what's our long-term backlog look like? And what are the ideas and things that we need to be putting into this product over time with this increased capacity? We've been in the process for the past few months is just reevaluating those roadmaps based on the assumption that possibly we will see this or likely we will see the throughput increase. I'm excited about the potential to increase this throughput. It's like a side benefit of all of the work we've done to move our customer base to our cloud. It's like now we have this -- we have a vehicle in the cloud-based installed base and the three releases we're doing every year to take the new functionality that we're building and put it in and…

Perfect. That's really insightful, Mike. And then, Jeff, just as a quick one for you. Can you just stack rank the areas of outperformance in the quarter as it relates to the ARR beat?

Yes, it's a good question. I mean I think in general, as I build my ARR model, there are the key elements that I need to see come to fruition. One is new deals in the quarter that then translated to ARR. The next is how much ARR is going to come off of the backlog. And the third is how much attrition events occurred. And we have really good visibility into the ARR that comes off of the backlog. We have really good visibility into those attrition events. And so those both performed largely in line with expectations. And then -- so then it's the new sales activity that we executed and delivered in the quarter is what drove that outperformance. A little bit of that we kind of called out was also some -- a bit higher true-up activity, but most of it was just the deal volume in the quarter, and then how that deal volume translated into year 1 ARR. Now within that, I think we saw a very healthy mix of kind of new customer wins, migrations, expansions into new areas within existing customers. And so that new sales momentum was pretty broad-based.

The other dimension to look at is geographical. So geographical line of business. So good spread across personal lines and commercial lines, which we're happy that continues to be a nice balance for us. The team in Europe continues to drive really solid activity and influence in the market showing up every day in the culture and the business of the countries that make up Europe and the U.K. And then our Asia Pac business continues. We were in Sydney last week with a lot of customers, and I'll just go back to the ProNavigator question. The receptivity, so many of those customers have -- are in the process of or already on cloud. Therefore, their appetite for consumption is just really -- it's a really powerful conversation. And so the Asia Pacific team continues to drive, I think, really solid market activity as we build out that leadership team, and we're really seeing that connection get stronger every quarter.

Okay. Great. We have time for a couple of more questions here. Next is Aaron Kimson from Citizens.

First one, there are about 90 Tier 1 P&C insurers today. Guidewire has 96 customers with fully ramped ARR greater than $5 million. How should we think about how many of your customers exceeding $5 million in fully ramped ARR today are Tier 1s? And how far down that TAM pyramid on Slide 5, do you actually have $5 million-plus FR ARR customers today?

Yes. I mean, I'll be honest, I haven't actually sliced it that particular way, but it is not -- it's very reasonable for us to have a number of Tier 2 and even Tier 3 customers that can cross that threshold. So that opportunity to see customers cross over that threshold is maybe broader than you might think.

Okay. That's helpful. Yes. And then, Mike, you mentioned strength with the analytics products. In F 3Q '25, you made your first Industry Intel sale within ClaimCenter. Can you provide an update on what you're seeing with Industry Intel, both from the standpoint of developing and validating models for more types of lines? And then also what John and team are seeing on the distribution side with Industry Intel?

Yes. We continue to make solid progress there. It's a process for -- and it's a little bit of a -- it's not a straightforward software development process. There's a little bit of have an idea about what we might be able to predict, go make sure that we can pull the data sets and clean the data sets and test whether or not there's appropriate signal that's in the data set and then validate that. And so there's a little -- it's a little bit more R&D and research than straightforward software engineering. But we continue to make great progress and steadily building momentum with that team. And so we're very, very happy. I would say we didn't call it out specifically, but sales momentum in the quarter continues to track as expected for the objectives on that team. And I'm very, very happy with that. So it's steadily building, and we continue to be happy with the progress.

Yes, I'll just add from a market coverage and distribution standpoint, the ability to demonstrate what that team has built has really crystallized quite a bit over the last couple of quarters. So it really helps in the deal motion. The other side of it is we continue to invest in our account management motion. And when these -- when the Industry Intel deals aren't necessarily tied to a large deal event, we're getting much better at navigating the right buyers inside of our existing customers and now with the demonstrability of those assets to have the right conversations to trigger a much more healthy pipeline activity into existing customers.

Our last question comes from Faith Brunner at William Blair.

I know there's a lot of commentary on the pipeline in the back half of the year. But just wanted to touch on maybe how should we think about the different products flowing through the funnel as customers increasingly want to land larger with longer duration. Has there been any shift to the conversations you guys are having or typical sales cycle timelines as people seem to be more eager to standardize on the platform?

It's an interesting question. I would say, and I'd love for John to comment on this is what we're seeing that's driving the outperformance is more like broad-based larger deals across the board rather than any sort of product mix shift that you might be thinking about. We're just getting basically much more established and establishing much more confidence in this platform as the logical long-term home for core system operations at insurance companies. The AI story, like we talked about, is driving some urgency there and bringing this to the table. But I would really say like the improvement is about larger, longer-term deals rather than product mix shift. We still -- obviously, we called it out. We still see the product mix shift, but like that's not what -- it's really core system, larger core system wins and commitments that's driving the improved momentum. Anything to add, John?

Time and stage is the same as it was, but we're starting now as we build out our portfolio, some of our portfolio will have a very different stage aging profile than the core processing space.

Okay. Well, thanks, everybody. It was obviously a great quarter. We're incredibly excited about it and look forward to talking to you all over the next few weeks and months. Otherwise, we'll see you at the end of Q3.

Thank you.