Roper Technologies Inc (ROP) August 2026 Conference Transcript

Good morning, everyone. Welcome to the Oppenheimer Technology Conference. Ken Wong here. I think most of you guys know me, software analyst. Happy to have with us the team from Roper. I've got Jason Conley, EVP and CFO, Zach Moxie, VP of Investor Relations. Good morning, guys. Welcome aboard. Good morning. Good to see you, Ken. Great to see you guys also. And for the audience, it's going to be a fireside chat presentation here. We do have the option to submit questions into the portal. I will pop those open periodically and take your questions. So kind of cue those up in your head and feel free to jot those down and shoot them into the queue. So with that, Jason, welcome aboard. Look, I think a lot of folks now, especially in the tech world, are generally very much aware of Roper. You guys have made a fantastic transformation from an industrial company to a software company. But even still, I think it'd be great to provide the audience with a kind of a quick background on Roper, and then we can dive into the formal fireside.

So, yeah, Roper is a vertical market software and technology company. and we're focused on sustainably compounding free cash flow per share in the mid-teens over a long period of time. We are kind of an end of one in the software space because we have this sustainable M&A motion, so we have a lot of M&A optionality. With that, we own 29 businesses. They're leaders in their niche vertical markets and typically we choose fairly smaller TAMs. We think they're attractive because of their protective nature and with our market leadership And especially with AI, this provides multiple paths to growth. So today, I think organically, we're sort of mid-single-digit plus, and that converts to sort of high single-digit cash flow just because of the margin and the low CapEx and working capital intensities is very low. So we take that cash flow, and then we just use a little bit of investment-grade leverage to acquire first call-on capitals, bolt-ons for our businesses because they just have such great returns. or, of course, buying the next great vertical leader. And so with this, you get this sort of continuous growth flywheel. And then, you know, obviously in the last three quarters, we've been much more active on the share repurchase front, just given we think are really attractive valuations at Roper. And I'd say I'd call it a somewhat paralyzed private equity market for the last several years and increasingly over the last six months because of everything that's happened with the public market. So, you know, we have a proven track record of acquiring great businesses and for reasonable valuations, and we make them better over time. And I think we're increasingly harnessing the collective learnings and value of those learnings and best practice methods across the portfolio, anything from, you know, AI product acceleration to what we call the product operating model and deploying that to continuous improvement methods and even into commercial excellence. So, you know, I think we're still getting going on how we think we can increase the organic growth across the portfolio and also just capturing more value from M&A. And so that's sort of Roper in a nutshell.

Fantastic. And actually, Jason, I'd love to start on that last point you made, the increasing organic growth. I think when you guys went down this journey, you guys were buying more mature software companies. You guys were probably sitting, let's say, more like low, mid-single-digit organic growth software companies. And now you guys have leaned into perhaps earlier stage software companies. Perhaps just give us a sense of why that shift, kind of where we are on that journey. You know, we'd love to get an update on that particular pivot.

Yeah, sure. And so, you know, about three years ago, we decidedly said, you know, it's an opportunity for us to look at businesses that were maybe think of first term private equity that had, you know, sort of a faster current of growth. that we could capture and help professionalize and mature that business under our ownership, you know, build a bigger platform, right, put on bolt-ons that make strategic sense for that business because we have a permanent ownership versus maybe buying later cycle where some bolt-ons were done, maybe they weren't great for the long term and sort of you inherit that and sort of have to work through that. So finding businesses that have faster growth currents and that we can help them continue to grow and then capture margin opportunity as, you know, as they scale, not cost takeout, but just kind of scaling as the business grows. And so, you know, the last couple of years, we've acquired Central Reach and SubSplash. We've done some interesting bolt-ons for our DAT business in a business called Convoy and also OutGo. And, you know, we've put together a much tighter governance structure around that. We have a value creation thesis when we're looking at a deal and then a value creation plan. And, you know, I'm happy to report the platform deals of Centuries and SubSplash have performed against our VCP in the first year. They're tracking against, you know, our forecast on revenue and EBITDA. It's just been great to have this tighter governance coupled with greater collaboration with our management teams. And so, you know, we built on the lessons of some of the earlier deals that we did, especially with ProCare. And so these are tracking quite well. And then, you know, the bolt-ons are continuing to be a motion for us. So we've invested in a team here to go and source deals for our, help source our deals for our businesses and collaborate with our businesses to be much more proactive and do more reach outs. And so, you know, some of the product of that is like our DAT business where we've now acquired a technology called Convoy. We bought that. It was owned by a broker at some point and then ultimately came into our hands. And so we're essentially automating the spot freight market instead of it being 10 calls between a broker and a carrier. We're trying to reduce that down to an automatic freight match ultimately. And so we're pleased with that. We've also bought a factoring technology business, and both of those are tracking well. It's going to take time. You're developing a new market, so it's a little bit of a wider range of outcomes in terms of timing. But we really like the dynamic kind of work that the DAT business is doing to create demand generation, to build that network, and then be sort of iterative on the tech. And so they're moving at pace. It's going to help our organic growth a little bit in the second half as that rolls organic. So I just say broadly, like the buying faster growth businesses, it's obviously good for organic growth, but also allows us to just capture more value for shareholders.

Understood. Maybe shifting gears from the broader strategy to more recent results, you guys recently delivered a very positive second quarter, improved your outlook for the year. I guess as you look back on the first half, how would you say that progressed relative to your expectations at the start of the year?

What were some of the underlying factors that drove that improved outlook yeah sure um yeah so we started the year our initial guidance was 2130 to 2155 and now we're at um you know we're at 2215 to 2230 so you know we had the obviously the buyback was a was a component of that so that plus better operating um performance has enabled you know four percent raise at the midpoint um i think operationally we've been very pleased with the first half progression um you know growth is particular in our our neptune business you know We sort of had a more sort of took a cautious approach given where that business was in its cycle relative to kind of COVID demand and how that was turning through. But they've actually executed very well in the first half. So they helped drive some of the outperformance at our technology enabled product segment. And it's just been better than expected. So this allowed us to raise our organic growth from five to six percent to now six percent. Also, I'd say, you know, software's performed well in line with expectations and a little bit better at our DAT business. We're finally starting to see improvement after, you know, three to four year freight recession. So what you're seeing is carriers are, you know, freight spot freight rates are better. Carriers are now coming into the market. So we stay with steadily seeing truckers come back into the market. And of course, there's subscribers entering our network. So that's been good. Our Dell Tech business on the private sector side, we think architect engineers construction, that's been going well for the last several years. But our government contracting business has been slow. I would say second quarter, we saw some signs of life with a large license deal that got through that we didn't have in our forecast but actually made it in. and the pipeline for Dell Tech looks strong, but we're not really ready to call a recovery yet just because we're sort of second and third order demand derivative in government contracting spend. So indications look good, pipelines, appropriations, and the like, but we're just going to kind of take a wait-and-see approach on Dell Tech. But overall, feel good about how we finish the first half and then how we roll into the second half as well.

Got it. And in the back half, you have baked in an acceleration in organic growth. I guess, what are some of the factors that are driving that? How much of that is within your control? How much of that requires further execution to potentially hit those marks?

Yeah, I would say mechanics play a key role in the second half acceleration. So I mentioned sensor reach and says flash. So Central Reach is in our application software segment. That turns organic in the third quarter. Like I said, it's growing north of 20%. It's in line with our value creation plan. A lot of that's recurring, so feel good about the security of that. And then our sub-slash business is in our network segment, and that turns organic in the fourth quarter and, again, growing above the segment average, so that'll add to organic. um and then in our our tep segment um our product segment uh we forecasted that to grow high single digits in the second half and that's especially and it'll be higher than that in the third quarter um our neptune business was challenged last year with some tariff and some copper headwinds that we are comping against so feel good about the the comp there but also just uh some of the fundamentals that are happening in neptune with a little bit more service work that's in backlog and just some of the mix of more static versus mechanical meter. So that's going well. And then I'd say just lastly, kind of a little less mechanic, but also a little mechanic, which is our DAT first half performance should continue in the second half. So just think of like a June exit sort of rate, how that carries into the second half. We're not assuming right now, we're not assuming carrier counts get much better, but just the natural June rate carrying into the second half helps us with that acceleration as well.

Got it. And shifting gears again, I think as you guys are aware, I feel like everyone in the audience is aware, AI has been a kind of for good or bad, a big topic within the software ecosystem. We'd love to hear what Roper is seeing out there.

What is Roper doing to address this ridiculously fast pace of innovation um yeah and and yeah we'll start there and there's a few branches to yeah of course branches um look it so you know i give i we really were thinking about ai as far as two years ago we really took this as a real opportunity for roper and so it all started with kind of the journeys of learning and and sharing of best practices which we've always done um bringing in thought leadership and really just getting at the mindset was kind of the first part of this and the urgency, which I think Neil did a good job of cascading that through all constituents, you know, presidents to the board members, to all of us. And so that really got us the compounding of learning starting early. And part of that was that every business had to kind of reimagine what their markets and their business would look like in an AI world. And that was as far as a year and a half ago. So we really got the, I'd say the juices flowing on that and the experimentation going earlier than most. We got agreements across all the frontier models a year and a half ago. So we've been at that for a while in terms of experimentation. And then we really leaned in in the third quarter of last year and hired some leaders to lead our AI function. And that's been a real, I'd say, turbocharge to the learnings and also to the actual product development so we're up to 20 people now uh as we sit you know kind of halfway through the year we plan to continue to add um and um we've been it's just been very successful the interesting part is the folks are attracted to roper because they um you know we're allowing there's not a lot of politics there's not a lot of boundaries um that we put up in blockers so these folks are getting to ship products they're like addicted to shipping products and they get to they get to context switch, right? So they go from Vertifor to, you know, PowerPlan to these other just smaller businesses where they can make a real impact. So we've had a lot of success with that opportunity and to work with a really high impact team. And the chemistry, I think, between the AI team and our businesses has been good. Obviously, you have bumps along the road because you're moving at such a faster pace, but I'd say overall, like, you know, early days, The experiment that we've done there has worked incredibly well.

Got it. And can you talk about kind of which verticals you're starting to see some early successes as far as adoption? And I realize everyone wants to put the monetization cart before the horse, but any color you can give as far as kind of that particular journey and how we should think about that path to revenue. Yeah.

Well, and I'll kind of break it up. One is our, you know, we acquired Central Reach in the second quarter of last year, and they were in the really early, not I would say really early stages, but, you know, fairly early stage of bringing products to market. And so we had about, call it, I don't know, four or five months of actual revenue when we bought the business of AI revenue. And what we observed there, there's a lot of learnings we observed there in terms of how you, you know, how you deploy products into the market, how you observe what the uptake is, how you sort of, you know, can do a freemium model that then cuts off and turns into something you can charge. And that's been a nice learning across the portfolio in terms of how to think about it. But so they've been really successful in being able to charge in addition to, you know, the core enterprise health record, just because it's a demonstrable value in terms of what they do for therapists and what they're able to do in terms of claims accuracy. And so, and that's in a market where you think about the, we've talked about the dynamics, it's in the autism space where the demand is so much higher than the supply. And so you're trying to get therapists to be as productive as possible. And also to actually enjoy their jobs, they're not doing a lot of administrative work because the turnover is really high. It's really, it's a really tough job at the front line there. And so that's what our solutions do. And so when you have that sort of that real need and you're able to fit it into the workflow, you get immediate uptake. I'd say in other areas, like I mentioned DAT, what we're doing is try to automate the spot freight market. Well, you think about a broker. There's an art in their mind for in terms of how you negotiate different lanes, you know, for a load. But there's always going to be a subsection of those loads that are going to be much more homogeneous, and those need to be automated. So you have to go through the change management with those customers, and there's a journey that goes along with that. So adoption is relatively lower there than, say, a central reach. And then there's all kinds of, you can paint a broad brush in between, depending on the use case of the industry that they're in and the urgency that's needed at the time. But I'd say over time, the solutions that we are creating are going to make the competitive intensity higher in terms of productivity in those industries. And once you get some of the first users, then the other customers will say, I got to do that too to remain competitive. So we think that it's going to happen. It's just, as you know, it's it's you're changing the entire way of working. And that just takes that takes time. And so we're part of the learning, too, is how you sort of redeploy some of your historical implementation folks to sit side by side with customers and help sort of increase that adoption.

Understood. A common question I get from folks is, like, you guys run a decentralized portfolio. Do you find that to be an advantage? Is that an issue as far as kind of getting everybody on the same page? What processes have you guys put in place to make sure that all your 29 businesses view this as a strategic initiative.

Yeah, and I think it's more than an initiative. I'd say it's a strategic pivot in terms of, and not pivoting from being a market leader and being intimate with the customer, but in terms of how you develop software, like that is, we've done a student body ride on full agent decoding. Now, it takes time to move that through the portfolio, but every single business is committed to doing that by the end of the year. And that's sort of not facilitated, but insured through our group executives who make that commitment. And again, it's all about sort of materiality. It's all about impact. So you don't have to boil the ocean all at once. We kind of look at the opportunity and then the risk around AI, and then we can phase these things in. But I actually think because we see so many different use cases, we see so many failures that happen in various ways to develop software. This didn't work. We learned from this. Don't do that. We actually, you know, it's an advantage for us because we are part of our job is to harness. Like I said, when I started today, we're really focused on harnessing the learnings and best practices and methods. And we're much more intentional about proliferating that out to the portfolio in a very structured way. So I do think it's actually an advantage, especially when you're dealing with something so nascent as AI. I think it gives us an advantage.

Got it. And then one more AI question here, and then I'll shift back to some other topics. um the competitive landscape i think there's again a lot of fear out in the market that software is being disrupted um i think there's a perception that you know you guys buy you know like strong verticals which should be a little insulated but you know a little more nichey so is it easier to attack you know so we'd love to understand whether or not you've seen any shift in the competitive environment from AI-native startups? You know, kind of where do you guys feel you may or may not need to kind of fortify the walls, the moat, whatever, what have you?

Yeah, I mean, we really have, I mean, we haven't seen anything of like significant threat at all. And we've got not just us, obviously our businesses are keenly aware of this. Think about it, they're the leader in their markets. They're close to their customers. Their customers provide them feedback if they're seeing anything. So we have had a couple of instances where point solutions have come in, not part of the full workflow, but something that our business didn't do. And we were able to replicate that in a matter of weeks and kind of shut out that, you know, that that competitor or that potential startup just because we're so embedded in the workflow. So I think that's still holding true. And, you know, there's the only area that we have one business that that that repackages public company information. And it's a data business. And it's part of it's it's part of one of our businesses. We have seen some AI native startups there, but they're really nipping at the low end of the customers. They've always had that sort of risk even before AI. So I think if you have a solution that is just pulling public information, you know, you've got more risk there, but it's a super small part of our exposure. So, yeah, we're not, and I think the startups are kind of going after, you know, obviously bigger TAMs now. And so we're working like heck to box that out. I think it's, to me, I think the work we're doing on AI is as much defensive as offensive. If you think about what Vertifor has done and putting out several agents already, like the customer is going, okay, I'd rather stick with one vendor as long as you're delighting me and doing the things that the startup can do. And we've got all of that proprietary data and sort of the entropy, the uniqueness of each one of those customers. We can create solutions that are unique to their workflow, which I think is sort of hard to replicate.

Understood. Makes a ton of sense. And I'm perhaps shifting over to sort of the other half of what you have to do besides kind of making sure all these businesses are operating in a very strong fashion. There's the kind of the pipeline of acquisitions. You mentioned earlier that perhaps the market was paralyzed, but you also more recently have indicated seeing a looser M&A environment. We'd love to dig into that comment. What are you seeing out there? What gives you confidence that perhaps things are starting to pick up on the M&A side?

Yeah, I mean, we've been wrong for three years, so with great humility that I am not a good predictor of this. But I would just say that we have obviously a lot of contact with sponsors, and so it's the bespoke conversations and the tone of those conversations with sponsors directly that give us some indication that they're going there because they're sort of reckoning with the current realities, be it cost of capital is the first thing, and then sort of what's happening in the public markets for software. So that's probably the strongest signal we get. And then just I would say that investment bankers are also giving indications that their pipelines are filling up and processes are starting to kick off. And then the third, I would just say, is the commercial diligence firms out there are quite sold out right now. Now, on those last two, the bankers and the firm, we saw that before, but I think these conversations with sponsors, again, they're much more constructive. They're much more sort of, I don't want to say off the record, but just kind of like more like let's have a real conversation versus like let's sort of they're trying to kind of test things out. I think it's more like, okay, let's see if we can get something done. So that's encouraging.

Got it. And has sort of this kind of AI landscape, has that changed how you guys go about sorting through which companies might make sense for you guys? Has that changed? Again, I'm not sure if there's a particular set of criteria that you guys are looking for. But, yeah, we'd love to understand how that might have shifted the mindset of your kind of potential targets.

Yeah, I mean, we've always been, I think, pretty good risk managers in general. Like, if there's some risk out there that's going to take the business to zero, we don't want to own it. And so this has just been, I think, to us, like kind of just an extension of that. I'd say we're acutely focused on AI risk and opportunities and deals. You know, on risk, we we've developed a moat scorecard that we sort of have initial hypotheses on deals. And then we we confirm it through diligence through a variety of ways. Right. Both on the commercial side and through technical methods, you know, with our team now, we're we're able to sort of really tease out some of the components of that. So and it's been it's a blessing, too, to have 21 software businesses where you can sort of compare, contrast and poke through all the sort of AI risk and opportunity in our businesses and use those as sort of parallels to the things you're looking at out in the market. So I think I think that's super helpful. And, you know, we always it's not only the business, but like I said earlier about kind of reimagining your market, like we were keenly aware of, like, how it will transform the way we're going. gets done in the end markets they're in, like a workflow that's there today might be something totally different in the future. So you have to kind of think through how agentic workflow will change that for better or worse. And then like how you assess if that's going to be a good long-term investment for you. So yeah, certainly we've, and it's been part of the build out of the AI team. We have these investment partners and we've been really focused on going through our portfolio in a very detailed way, kind of scoring those businesses and then using that same framework to then do diligence on businesses.

Got it. And as you touched on earlier, one of the better returns on your dollar right now has been to buy Roper stock. I guess, how are you thinking about kind of the use of capital going forward? Does it kind of shift back towards the M&A direction? Is Roper stock still attractive to you guys at these particular levels to be as aggressive as you have been?

Well, look, I think it's an opportunity cost conversation. And for all the reasons I've just described about the length of time that we haven't seen transactions, the cost of capital reality on the 21, 22 portfolios versus where it is today, what we've seen in the public markets, those all point to signs that we think is going to be a good, a rational private equity market for us to buy things at reasonable valuations. And so this is really about keeping our options open. We still think at these levels, ROPER is an attractive investment, but we also think that the potential opportunity for M&A is going to be great. So we are planning to, you know, sort of do the pause on the share repurchase for now. But, you know, look, if, like I said, I'm a terrible forecaster. So if a year from now, you know, things aren't transacting and we're still in this, you know, kind of AI valley of despair in terms of perception, we'll continue to lean into, you know, to the buyback if that's, you know, because we'll have a lot of capacity by then.

Got it. Now shifting back to AI again, so we touched on a lot of the competitive dynamic, the top line dynamic, but I think the other piece of the equation has been like how this may or may not impact your margins, right? So you can execute it fantastically and you drive a lot of AI revenue, but then there's sort of a flip side to it, which is, okay, the margins are always going to be a little lower than SaaS because of the inference costs. So what are you seeing as far as margins? How should we be thinking about that going forward? What kind of policies do you have in place to potentially prevent a lot of these kind of token maxing headlines that we're seeing from some of your other tech software peers?

Yeah, well, so I'll handle kind of the gross margin first, then we can talk about the token maxing. I think once you get – just what we've observed so far, so take it for kind of – we're still early days. But I think when you develop AI agents, there are certain ways to codify certain rules and not have to use inferencing on every transaction, right? It kind of depends on the thing that you're doing. so that actually can increase your gross margin and then it really gets down to like what exactly do you need to call a model for for what specific kind of subtask and i guess our experience is your gross margins may start off worse than sass but you can you can work that down over time as you have more experience uh and you have um that sort of better methods or cheaper of course cheaper models we we don't use any of the frontier models for our products we use them to develop them and to do sort of exploratory work but once it's into production there's there's really no need um you know to to use the because we're not we're not doing any this is just this is task work and it's a agentic task work it's not it's not rocket science so i i think that's that's what we're observing and so i and i don't based on the nature of the things that we do for our customers I can't imagine us having to use a frontier model for some basic day-to-day tasks. I just can't picture that. And I do think at the lower end, as you know, the model costs are getting, like the older generations are getting much cheaper. And I'd say even things like how you, the types of prompts you do, like caching or batching, you can get your costs down there too. So it's not just the types of models, but then the types of prompts you do. So a lot of that goes into it. So long story short, I don't see it today as having a gross margin challenge with AI products. Now, as compute goes up and it gets more scarce, you know, those are the conversations we're having about how do we think about broadly across the portfolio? Do we need to think about buying ahead for compute? And those conversations are happening so that we can make sure that we're servicing our customers. We don't want to get into this situation where we don't have the capacity. But that's just kind of thinking ahead as, you know, electricity and the compute becomes more scarce. On the token maxing, I think the, you know, we've gone up. So, like, just in terms of cost, we've gone up maybe 3x or so since January. where it's certainly going to continue to go up, you know, by the end of the year. But the good news is we've got, I think the local ownership really helps kind of balance both speed and control because every business is keenly aware of what their kind of budgets are and then thinking about routing, you know, certain prompts to certain models. So we're pushing out a lot of best practices around that. Some businesses are doing that auto routing. Some are using, you know, spend management controls and dashboards. We have visibility to almost close to having visibility across the portfolio around that. So we can think about not to micromanage them, but really, again, to think about the future, how we think about, you know, compute across the portfolio. So I'd say, you know, the hallmark of Roper is kind of that local sort of obsession with, you know, kind of local business activity, which actually is a strength for us. So we don't get into this like runaway token maxing because there's just so much accountability across the portfolio.

Got it. And then how are you guys driving efficiencies internally? Where are we on that journey?

I think we've seen some of your peers are already starting to extract a fairly meaningful margin expansion. um what about roper is this something that you guys anticipate being able to squeeze out incremental margins going forward as you guys implement ai more across your organization yeah i mean we're certainly seeing going back to agentic uh coding agentic development so we've got a three or four businesses that have completely moved to that we want to continue to push that like i said by the end of the year every business is committed to to doing that we do think there's clearly a lot of productivity that we're seeing orders of magnitude um at the developer level our goal though i would say this so we're not looking to um to expand margins dramatically as a result of that we're looking to take that put it back into the roadmap to continue to just have that flywheel of continuous innovation so that we can inflect continue to inflect organic growth. There may be a point where the productivity is, you know, sort of outpaces that level of innovation, but I don't think we're anywhere near that in terms of the opportunity that's in front of us. And I'd say mostly, like most companies, it's primarily in, you know, in our to what i said earlier like the two years ago the when the when the gun went off or whatever we were really more focused on revenue because it's so much harder getting the efficiencies um not not easy by any means but certainly a more commoditized you know um you know um skill to to to develop over time got it and and then uh the the last piece on the ai side and something i'm sure all

software vendors are trying to figure out is is the licensing piece because how are you guys thinking about licensing on your end uh our customers well i don't imagine customers are are looking to necessarily change the consumption base but but yeah basically we'd love to understand how you guys are blending those particular models do you see a situation where down the line you will have to pivot to more of a usage token-based model yeah it's early i mean i think we so like with vertifor um and some others like they're we're building a platform of agents and it's going to be more of a credit model so they can sort of as they're using the agents there will be a drawdown

on that so think of it can as like recurring revenue but it has some form of consumption and if there's overages there'll be a a true of after the period's over maybe it gets rolled into the next period you know there's probably some flexibility around that so so that's like one end of the spectrum then you have like our software business so this is pharmacy automation software for long-term care pharmacies they have a sort of a an agent that helps with taking orders from long-term care facilities and getting into the system which sounds somewhat mundane but it's actually quite challenging and it would be able to sort of identify that that's more probably going to be transactional because that business a lot of that business actually has a lot of transactions already so the customers are used to it um then take another example of like you know our delltech business where we've said you know we're going end of support on um on our cost point solution in govcon in the first quarter of 28 we're trying to get customers into the cloud which obviously has a two to two and a half times lift and all the agentic features are going to be in that cost point solution in the cloud. So trying to kind of get the price through that method. So I say all this, and then we get to the central reach where the AI is based on per learner. So that's kind of a form of consumption, if you will. Like if you have a learner for six months, that's going to be the method of charging. So we get a good sampler platter across Roper, and we're just trying to educate all the folks. I'm actually creating a pricing agent from all of the different artifacts that we have to kind of share that best practice out to the businesses and considerations and different modeling modalities that you can do based on what we've seen at other companies. So I know it's a long-winded answer, but it's a little bit of everything.

Got it. And this last question of mine might be a long-winded, a little bit of everything as well, But now kind of bringing it all back to, let's say, more of the near-term operations. So we've got the beat and raise from Q2. I guess as we think about kind of where things can go right, where things can go wrong as we enter the back half, we'd love to get a sense for how we skew towards the upper end or the lower end of your guidance ranges.

Yeah, I mentioned Dell Tech, right? I mean, I think we're taking a somewhat cautious approach, rightly so, because you could have a lot of perpetual there. We don't know if it's going to land or not. We'd love for it to not be perpetual. That's the other wild card is best. And what I just said about cost point being into support, we could have more go to the cloud so it won't impact, you know, the current year as much as it will next year. So that's that I'd say that's still a wild card. um you know i think if you go through dat i talked about i think that's in good shape could get a little bit better if carrot you know if carriers continue to enter the market um but we'll see how that how that plays out um you know and so the good news we don't have a lot of right the range of outcomes for roper isn't that great so as we as we sit here today you know our most our most uh higher beta uh segment in terms of just growth is is is the tep segment if you go talk about quarter to quarter, but we feel decent about the comps in the third quarter. Fourth quarter is going to be a little bit lower probably, but, you know, they're not a high backlog business, so that can swing a little bit either way as we get through a quarter and get into the fourth. But that's sort of where I would, you know, look to in terms of variability relative to our forecast.

Understood. And I think with that, we're right up on time and I am at the end of my question list. So I kind of laid those out perfectly. You did great. I appreciate the conversation, Ken. Yeah, Jason, always happy to have you. Zach, really appreciate all the help you always give us. Thanks a lot for supporting Oppenheimer. Have a good one.