Thank you. Good morning, and thank you all for joining our Q2 call.

Before we get to questions, 2 things I wanted to put on the table upfront. The first is AI.

It is changing 2 things for us. Our customers are asking our software to do more for them while our businesses are finding new ways to build better products faster.

We are seeing both. What we are not doing is running an AI program out of head office.

Our business unit managers understand their verticals far better than we do, and they are making those calls themselves at their own pace, funded out of their own P&Ls. Some of that work will pay for itself and some of it will not.

And we will find out the same way we find out about everything else we fund, which is slowly. As many of our businesses have been adopting AI, we are seeing them use the tools to rapidly move through our customer-driven product road maps and augment our products with AI functionality.

Let me be direct about what that means to you. We are not going to give you an AI target, an AI revenue line or an AI time line.

If we start reporting a number like that, we will start managing to it. Our software is primarily mission-critical.

It is embedded in how a customer runs their businesses every day. AI is something we add to make that software more valuable.

The second thing we've been working on is verticalization. We have been grouping our businesses and our prospects into coherent verticals where possible.

The purpose is simple. We want to be the obvious permanent owner of a niche so that when a founder starts thinking about the next 20 years and selling their business, we are the first call.

That kind of reputation takes years to build. It will not show up in this quarter's results or next year's, but I would ask you not to look for it there.

The constraint is not structue, it is having the right leader for each vertical, and leaders of that caliber take a long time to develop. We are setting the pace accordingly.

The 2 are connected. Depth in a vertical is what lets you tell the difference between an AI product that matters to a customer and one that merely demonstrates well.

We would rather have that judgment sitting close to the customer and understanding their needs. So thank you very much, and I'll turn it over for questions.

I want to ask for an AI metric, but just qualitatively, as you look across the portfolio, are you seeing a growing number of examples of businesses that are starting to see an uptick in organic growth driven by AI product enhancements, the way we heard some of the case studies at the AGM? Is there more and more of that happening in the portfolio that you can speak to?

What we're seeing, Thanos, is we're seeing our development process is getting much better. They're moving faster through backlog.

So we're seeing a productivity increase, I think, across the group, not everywhere, but it's starting to adopt because we spend a lot of time working on training up the -- all of our resources on how to use AI tools better for development. And right now, they're working to talk to customers about potential add-ons or additions to their products that are using some of this functionality to develop the software, but we still haven't seen a real pickup in organic growth from it.

I think that's a ways out. what I always say, you can build products fast, but selling them is a whole other thing.

A customer has a budget for them and see -- and you have to be solving a need that they're willing to put some money on the table for before it's going to impact our organic growth. So I think that's consistent with what we've seen in previous quarters.

I'm just really happy to see our teams adopting the tools to develop software faster and more efficiently than ever before.

Great. And then we've seen some PE acquisitions you've done recently with DerbySoft, TouchBistro and Imagine.

Is that indicative of a broader trend with respect to valuations coming down on larger assets, or kind of more of a case-by-case thing?

I think it's the latter. It's Bernie here.

It's more on a case-by-case thing. And we're finding acquisition opportunities across the spectrum from owner managers to PE to carve-outs.

So those are still coming through. It just so happens that we had significant volume this past quarter and the beginning of the third quarter.

So we're still seeing all of these opportunities proliferate, but it takes a lot of time, and we've developed these relationships over the long term. It just so happens that some of them happened and just took place this quarter.

So we were very fortunate that we're the first ones that they called, and we managed to close a whole slew of them.

I just wanted to circle back on your comments around grouping businesses into verticals. It's a little bit different than CSU's historical focus on kind of not integrating these acquisitions.

Just curious if you could talk a little bit about what's changed in the market that's led you to evolve into a potentially more vertical market focus and whether the verticals will be between different operating groups or how you think about those verticals?

It's a good question, Stephanie. It's really -- we're not integrating businesses.

There's no business integration going on in it. We're just trying to make sure where there's a chance to, let's say, I like to use a Constellation expression, put them in the same orbit together, where they're not, if they're across operating groups, we'll move a few businesses in order to try to make sure that there's someone who oversees a particular vertical.

And it allows us to use some of these tools that we're seeing with AI that we might be able to use across some of those verticals. But that's the hypothesis.

Time will tell. We've had good success when we've had vertical-focused leaders inside of Constellation that have created, for example, a Lumine and things like that.

So I really do -- I'm just trying to refine that a little bit and get the businesses that should be closer together, closer together, but I do not plan on -- or we don't have any plans to integrate them in any way functionally or in any other way. But if they're a little closer together, they might be able to learn a little bit more from each other and it positions them better for the next 5 to 10 years.

Okay. That's good color.

And then maybe on Altera. So Altera's organic maintenance and recurring growth decelerated in the quarter.

Just curious if a large client decided not to renew like, how should we think about the Altera organic maintenance growth and Altera going forward?

Yes. I tried to put it in the MD&A.

They're a SaaS, but it was a very -- like in Q2 '25, they had a very strong quarter. So they actually had a couple of new name sales.

And the way IFRS makes us account for things now, if you have a large contract, you have to recognize a certain amount upfront. So they actually showed, I think, this positive 1% organic growth in Q2 of '25, yet the trend line for this business is going to be a slow shrinker for the next year or so or a couple of years, I would say.

And so that is the driver of it. So if you normalize for that strong Q2 '25, things are -- there's nothing terrible going on.

So I'd expect that organic growth to revert back on a full year basis to sort of what it has been trending at recently.

Just echoing on that last comment on the organic growth, Jamal. So even if you exclude Altera from the maintenance and other recurring revenue growth in constant currency, you came in at 4%, which is below the historical range of 5% to 6%.

So was there other stuff that you would call out, I guess, as it relates to the organic growth?

Yes. I mean there's a couple of other items like it's tough because you don't talk to specific BUs in the MD&A.

But we had a couple of larger acquisitions like Dark Matter where a similar thing, where in Q2 '25, they actually recorded organic growth of -- it was 10%, right? And again, this is a business that we're still fixing has negative organic growth right now, but you had this huge Q2 '25, and so therefore, you've got like a negative 18% in Q2 '26 because of that comp.

But the other thing, if you look at some recent acquisitions, so Lumine is broken out. You can see that they've made some large acquisitions recently that they're pulling out.

Their organic growth in the quarter was 1%, right? So again, a drag on CSI, but things that they're expecting to turn around.

And then there was another large -- or a business that we had in South America, where they lost a large customer. But this was an example of a business that we knew at the time we acquired it, that, that customer is leaving, has nothing to do with AI.

The customer has now left, but it was sizable. And I mean that customer alone was like a 30-basis-point drag on CSI's numbers, right?

So there's a few of these like one-offs that are causing it. Many -- or the 2 that I talked about, like the Altera and the Dark Matter are purely accounting related and should revert back next quarter.

The Lumine thing is like they're fixing these businesses. I don't know the exact time line of when they expect to turn around, but yes, I could -- if you back out those 3, 4 things, you could normalize back down to that sort of 5% number that we've always trended at.

No, I appreciate the color. And are there any of these kind of whether it's tougher year-over-year comps or maybe some customer -- larger customer attrition that we should be looking out for in the coming quarters?

Or does it look a little bit more normalized?

Like my expectation based on what we own today is it's normalized. Like we were also making a lot of large adaptations, and I have not analyzed those to see what their impact will be in the next quarter or so.

And -- but yes, based on what we own today, I would say, yes, I would expect these anomalies to sort of revert back. I'm not expecting another large customer leaving or anything based on what we have today.

Yes. That's helpful.

And maybe one question on Bernie. I think you kind of talked about some of the deal flows and more increased deal flows of, say, larger deals, $100 million, a few hundred million dollars.

So it sounds like you're seeing more of that. Are you also seeing better win rates for those deals?

Like how is the competition for them?

Yes, I think the competition is still very robust. No one is giving up on vertical software, whether it's large or small.

We're seeing some weaknesses at the high end, but -- in pricing, but it's still very, very competitive. So it's not like we're increasing our win rates or anything like that.

It's same old try to get as much as we can. I don't see any improvement, but we had so many people out there looking for acquisitions across the board or across the globe.

And it's a matter of developing those relationships over the long term.

Can you comment -- have you seen any changes within the kind of the competitors for the businesses, whether they're copycats, strategics, potential buyers?

The copycats are still out there. I don't see any new ones that are popping up really.

It's more of the ones that have popped up in the last 5 to 10 years, and those still exist, and they're still competing for businesses. And I expect that to continue for still quite some time until maybe some of those portfolios decide that it's time to move on and they don't want to go public and they just want to sell because it's PE backed and there's a limit to their funds.

So I don't see any changes there. I don't hear of new ones popping up or if there are, they're very small, and we hope to pick those up at some point in the future, if at all possible.

And hopefully, at some point in the future, the number of copycats starts to reduce, but we just don't -- we can't predict that. We don't know.

And one last question maybe for Mark. When you were talking about the verticalization, are you changing in terms of the M&A and the BD teams for that?

And how do you see an impact on potential movements in the portfolio for these -- or the companies within the pipeline for the personnel there?

Well, we've seen a lot of -- when we have vertical groups that are larger inside of Constellation, we've seen a lot of success deploying capital in those vertical groups. So definitely, some of the M&A resources will be better aligned to verticals than stepping outside of them in some cases.

So there definitely will be some changes related to that.

I was hoping, could you speak to the profitability of the 2026 cohort of acquisitions? Is it fairly typical and comparable to other cohorts?

Or do you see a change there?

Yes. I mean, I think I called it out last quarter that it was like negative margins, but there were a couple large -- it's more the purchase accounting, like there were some bonuses that we couldn't put as part of the purchase equation that we had expense, et cetera.

But that cohort in Q2 meant from being -- I think it was negative 16% in Q1 and it's now positive 16% in Q2. So it's -- and it's going to trend up.

The 2025 cohort, similar thing, like I think it started off around 16%, 17%, it's up to 20%. So it's moving up.

So -- but all the -- it's not that these businesses will continue that way, like it's just taking time to improve them to the levels, but if you look at all the cohorts prior to that, like they're probably in the 30s or high 20s, right? And so it's just taking longer for them to go there, but no, it's not -- they're not going to be inherently lower forever.

Okay. That's helpful.

Second question is just specifically on DerbySoft, I mean there's some disclosures in the MD&A. Is it fair to annualize the revenue -- the disclosed revenue in the MD&A and you get to roughly the $100 million and the purchase price is $400 million.

The multiple seems high. Is there anything wrong with my math, one?

And then secondly, is there any reason for the multiple being higher than typical?

You want to talk to the multiples?

Yes, it is a high multiple, but it's a very successful organization. It's growing nicely, good profitability.

It could always be a bit better, but that's what our expectations are of the business. It's a very solid business.

But you did use leverage as well, right? And so remember that helps us pay a little bit more, but...

We also have a great team at Constellation working on that one, Paul, as well.

And the punchline on the multiple is that you still expect the IRR to be in line with acquisitions of the similar size?

Absolutely.

Yes. Hurdles rates aren't changing.

Yes.

All I've got 2 topics, I'd like to ask a few questions on each. First is Altera.

And I just want to thank you for the ongoing disclosure for that asset. It's been very helpful, and I'm very appreciative of that to be able to track the progress on Altera.

So we've crossed the 4-year anniversary. We acquired it in May 2022 and since then, according to my math, that's generated a cumulative $400 million in free cash flow, all while revenues have declined from a little over $800 million to now about $646 million in trailing 12 months.

First question is, do you think these results have tracked with your initial expectations?

Yes. I mean I'm not closely involved with the business, but I have the CEO of Harris has talked to it at the Board meetings, and he has said, yes, it is aligned with sort of what they expected from the investment thesis, and it's aligned.

It's actually just operating ahead of what the expected IRR was in the original investment so...

Got it. And also on Altera, I was curious if anyone could share any general learnings that you've had since acquiring it?

What has surprised you good or bad since acquiring Altera from Allscripts? And have you been able to apply any of those learnings to acquisitions since then?

I would think generally just -- it's a very large acquisition and it's something we've been learning how to do better over the last 5, 6 years. We've done a bunch of larger acquisitions.

And one of the key learnings is driving best practices is harder to larger acquisitions. You can't drive them as fast.

So the ability to maybe break the business up into, let's say, what you would call smaller business units is something that we're trying to do faster when we can, when we acquire an above average sized acquisition for us. And so that's how quickly can you do that is really defines how quickly you can improve the business.

So -- but that's just ongoing. And that will probably be the biggest focus.

May be, Bernie, you want to add something to that. It's just driving best practices is harder in larger acquisitions.

The complexity of larger businesses is really the challenge. And the various groupings, different product lines, you have to manage them as a portfolio.

And because we do that on a portfolio basis in CSI, we have that experience already, but there's -- there are cultures that are just different from what we're used to at CSI and some things have to change, and it's just a little more difficult to change them.

Back when we were doing smaller acquisitions where they were sort of $5 million in size, $4 million or $5 million in size, it was really easy for one person to wrap the head around what had to be done to do that acquisition, how it should be improved, what best practices you could apply to it. When you're working with a much larger acquisition like an Altera or even a DerbySoft or any of these, you have to really be thinking about a team of people helping you think through what to do there and implement it, which you didn't have to do in the earlier days when you're doing smaller acquisitions.

One of the nice things about us is we've learned how to do that a lot better and more people are able to think that way than we would have, let's say, 10 and 15 years ago.

Yes. All right.

So second topic for me is continuing on the corporate carve-outs. So I guess, in general, I mean, like the difference between a small acquisition and a big acquisition, you have to change how you approach things and drive best practices.

But on carve-outs, is there any reason that they're a bit [indiscernible] deal with? Can you talk about those reasons?

Are there -- when it comes to analyzing it as an investment opportunity, what sorts of financial adjustments you have to make? With sort of qualitative adjustments you have to make?

Are there different risks? Are they greater or lesser and other types of risks?

I just appreciate some more thoughts there.

Yes. I mean we get better than every time we do them.

I mean, Bernie might want to just add -- Bernie, might want to expand on this, but they're interesting because it isn't -- they're not clearly -- all of their financial statements aren't fully separate out in a lot of cases. They're interwoven together.

You have to separate them from internal systems that they're on. There's -- yes, there's a whole bunch of carve-out things that really you have to manage carefully.

And plus whoever you're buying the carve-out from, they really care where their customers end up in a lot of cases, right? So you've also got to balance that as well.

So there's a lot of things you do differently with a carve-out that you wouldn't with a stand-alone acquisition. Bernie, what would you add to that?

The same goes for the way we look at businesses in terms of return on investment. And every decision that we make within our business is based on return on investment.

And if we can't see that return, we just won't do it unless it's really, really strategic as it were. And a lot of the businesses that we do get from carve-outs don't look at their businesses that way.

And so that's a bit of an infusion that we give them. And we try to set their minds towards measuring that investment.

And so -- and there are lots of decisions that these businesses make on a regular basis. And when it's a larger business with various product lines, you have to take that decision, make them more granular and infuse that kind of thinking into the people that make those decisions.

They're generally not very balance sheet driven, right? They're not thinking about their balance sheet and that's -- the balance sheet is obviously something pretty important to understand and just figuring out what that is and getting them to think about managing their working capital is harder than it would be for a standalone business because it's -- there's not -- usually not a separate balance sheet for that particular carve-out.

Awesome. Awesome.

Great. And just one final one on carve-outs.

Would you agree that in your universe of potential acquisition targets, are carve-outs typically an area where Constellation can deploy larger amounts of capital?

Yes. We've been doing it for years.

And I think that the fact that we've done it with Fortune 500 companies just underlines our ability to do carve-outs appropriately and maintain that customer base as expected. And I think these larger businesses are more than happy to deal with us when it comes to taking care of the customer base and the employee base.

So I think we'll see, hopefully, more of those.

And we have done multiple carve-outs from individual businesses, too, which is kind of nice. Your reputation is important, very important that you take care of their customers and their employees when you're taking over a business like that.

And that matters to them.

And can you put any kind of general number on potential carve-outs that might be out there? Are they in the hundreds?

Or is it greater than that?

No. I wish we could see the future, and no.

So yes.

Great. Nice results here.

It seems like the M&A, particularly for 6 months of the year have been elevated. Should we expect a similar pace in the back half of the year?

Or just any thoughts on the cadence just given there's been some pretty good success in the first half of the year?

I hate to predict things. We have a very robust funnel.

We're working with a lot of businesses, whether they happen in the second half of the year, whether they happen in the future is impossible to tell. You've seen our results last year were lower.

So it's just so hard to predict when these businesses will actually close with us and whether they will close. So I will not offer a prediction.

We just want to make sure on the playing fields when they're available, right? We want to be out there trying to win.

Super helpful. And then the pacing of the organic growth.

Obviously, last year, the first quarter, there was only about $94 million of acquisitions as opposed to $380 million in the second quarter. Does that $380 million start to come into the base in the third quarter?

Said another way, right, is it the third quarter where it starts to get into the organic base?

Yes. The way I calculate organic growth is I add my pro forma, our history as if we own that business for the full year or for the prior comparable period.

So if we're buying businesses that are turnarounds and we need to fix or whatever, it impacts organic growth right away. And that's why I do it that way as opposed to waiting a year or whatever before it starts showing in our numbers so...

And that's kind of where it was a little bit slower because some of those ones...

Yes.

Got it. No, that makes a lot of sense.

And then 2 others, and I'll get back in the queue. It seems like the taxes in the second quarter were a lot higher than what we modeled.

Was there anything there just relative to the income taxes paid in the second quarter relative to -- it was almost double what we thought...

So if you're looking at cash tax, like I mean, that's just timing and -- but I always look at the current tax expense as a percentage of like sort of this adjusted net income before tax, and that number is sort of in line in that sort of mid-20s, which it always is. And yes, so current tax approximates cash tax.

If you look at the current tax line, it's pretty stable. The cash tax will fluctuate based on payments.

Helpful. Last one for me.

The commentary on AI was super helpful. As you're thinking about implementation internally from an expense perspective, any way to think about where you are in that process?

And then any way to dimensionalize the tokenization side of it in terms of increasing token costs relative to productivity? And how are you just managing that?

And is there any way to -- again, is the token cost where you thought it was going to be? Or just any thoughts around that as well?

Yes, we started tracking this. So we've created now GL accounts that will track sort of the token expense and any cost related to AI.

I mean, still early days. I mean, if you look at our P&L right now, you'll see third-party maintenance like slightly up, but nothing material.

I'd be honest with you, I don't have a great, a clear number of what these total expenses are relating to AI. So this is something we'll start tracking now going forward.

But I do not believe it's materially impacting us today. And it is something that -- as we talk about this internally, that we will -- it's something we take into consideration even like the tools that we use, making sure we're not beholden to any one provider, and yes.

I agree with Jamal. That seems correct.

I think in some cases, we -- you're going to -- where we push people really hard to start using AI tools to start increasing their productivity and ability to get customers what they need faster. And I think it's really interesting to see where the tools evolve.

In some cases, we're using -- one of the expressions internally is we're using a blowtorch to light a cigarette. So we're using very powerful tools for problems that probably don't need those today.

But it's just -- you sort of want to, I would say, role with it right now in order to make sure you're adopting technology fast, people understand what's possible. So it's going to be an interesting journey for us.

The great news is we haven't banked on any one particular platform to use. So we'll learn from each other and adapt as necessary.

Our businesses will have the opportunity to very quickly change courses if -- course and what tools to use if they need if we're thinking things getting to expensive. So...

This is Teddy on for Samad Samana. Congrats on another solid quarter.

A handful of software companies have called out headcount this earnings season even with plans to reduce or aim for flat headcount, including a shift away from G&A towards sales and marketing. How does that compare with the thinking across your portfolio companies for the balance of 2026?

No, we're not taking really a lot about headcount reductions. I mean, we have a lot of -- remember, we're a very decentralized organization.

We don't have this big 20,000-person R&D group that works across the world. We have individual businesses all over the world that have development groups that, in some cases, have single-digit number of people, and in some cases, have double-digit numbers of people, very rarely have hundreds of developers in any one business.

And so we're really trying to get them to be able to offer their customers more faster and get through product road maps and fix bugs and things like that using these tools. And the headcount reduction is not a focus of the company at this point.

Thanks so much for doing these quarterly calls and the transparency. It's really wonderful.

That's Barca Capital. Sorry, I don't know...

My question is not about AI. Based on public information, it seems like a lot of your IRR when you buy something comes from changes in working capital.

I am just curious, is this generally the case? And if so, could you give a very rough sense maybe of what portion of the IRR comes from changes in working capital when you buy something?

The reason that it's important is, if people are thinking about cash flows from your deals very long term, if a lot of the cash flows are showing up earlier, it doesn't matter that much than if they're showing up, I don't know, in 15 years or something like that. The other question I have is not about AI, but it's just general.

It is, have you seen a change in the number and profitability of your [ capital AI ] initiatives in the last, say, 6 months or 12 months compared to 2 or 3 years? Or is it about the same?

Have you seen profitability much, much higher, much, much lower? I don't know you're doing 5x of these or 10% more, or it's just about the same?

Again, thanks for doing the call. We all appreciate it.

We appreciate the tone, the transparency and the disclosure.

Okay. So just to hit the working capital question.

Every acquisition is different. I know back in the early days, like the very early days of Constellation, it was a big factor in our acquisitions, but it didn't mean that it detracted from the IRR of the businesses themselves, or the cash flow that we could generate from those businesses.

Today, again, working capital changes make a little bit of an impact, but not a lot. The biggest impact is from running the businesses appropriately and getting the cash flows out of those businesses.

It's really operating changes that we make to the businesses as opposed to working capital. And if there is an added bonus that we can find within working capital then all the better.

But I wouldn't say that it's a significant portion of our returns. I hope that helps...

Is it fair to say, or have you had an eyeball on trying to get those cash flows? I know you're very time, value, money oriented, but getting those cash flows earlier rather than later or saying [indiscernible] 4 or 5 years, we don't think too much about it?

Sure. We always try, but sometimes these things do take time.

In Europe, I think it takes maybe a bit more timing than it does in North America to make changes to our businesses. But we try to get them done as quickly as possible.

That is our objective to make sure these businesses are operating appropriately like our 1,500 other -- plus other businesses so that they're in line with our operating guidelines. And so we do try to get them done as early as possible, yes.

Sometimes you have to look at customer contracts, and it just takes time, right, a long time, especially with large customer contracts. Those relationships are very important as well.

So you have to think through how to make those, but we model that all up as we think doing adjustment carefully. So...

But operating cash flows is first and foremost and way more than working capital. And if we find those opportunities with working capital, we definitely do take advantage of those, yes.

We measure it very closely, very closely. You were asking about initiatives, I think, as well.

So yes, I mean there's a lot of initiatives going on throughout the organization. There's sort of 2 types of approaches to initiatives.

We really love customer-driven initiatives where the customers actually say they need something, and we help them figure out what that is and some businesses that are incredibly driven by just what the customers want now, especially our larger customers we do things for. And then others are creating -- like there's a number of AI initiatives going on, for example, right now, where they're creating potential products by listening to what we call revenue signals from customers.

So I think there's probably more initiatives now than there was a year ago. And we'll just have to see what those amount to in the end, what the actual returns on the capital that we're investing in doing those.

So we have a very open mind to people experimenting right now with the technology evolving and AI.

Do you have anything you could give us a sense of how well those early ones have been doing?

It's too early. It really is.

It's really too early. There's some interesting -- what's interesting to me like if we're seeing someone -- what we've seen, I think, this quarter, if we see some other outside company coming into potentially to sort of more of a horizontal play in some of our verticals, we are able to be much faster followers, which I hadn't thought a lot about.

We've always had the ability to be a fast follower, but you can be a faster follower with AI, especially when you have a lot of customers in that particular area. So as long as you've skilled up your team so that they're able to move fast, they can move much, much faster and provide customers something.

So that's kind of a neat outcome, right? Because if it took years to build some products like some initiatives, it takes 7, 8, 9 years to get them to the maximum revenues over time, believe it or not.

So -- in the early days. So maybe some opportunity to do some things faster in certain areas now where other people have driven up the, what could you say, the desire for that type of a product line.

So it's fun to watch that for me over the next few years.

That's great. I might sneak in one small question.

You operate in many different countries and many regulated industries in particular. I think you said that regulated industries are often less likely to prefer cloud solutions or maybe AI solutions.

Have you found that country-to-country regulated industries have different preferences with respect to on-premise or AI coded products?

It's more sector to sector within countries, I'd say. I mean, I don't -- I wouldn't have a broad conclusion on that.

I'd say it depends on the sector. If you're government, you're clearly very concerned about those things, bank, banking, like health care.

So it's sector-specific in every country, I would think.

I've got 2 more questions. I'm curious if you guys, the management team and Board, are still studying high-performing conglomerates and other successful businesses?

And along with that question, I'm just curious, how did those learnings trickle down? Or how far did they trickle down?

Is it just the top people that are studying these other businesses? Or are there hundreds of business unit managers that are also trying to glean lessons from these case studies?

I think -- it's a really good question. I mean, I've been fortunate enough be able to be involved in understanding that as well as the senior team at Constellation.

I think it really varies by the operating group how far they drive those down with inside of Constellation. There's a lot of content that we shared for people who are -- as we're onboarding them, as we're onboarding new companies on how we see the world.

And we did -- Larry Cunningham did a study for the group that I oversaw for many years, Volaris, and he went and he spoke to -- interviewed 70 of our leaders across the world and was -- really wanted to understand how invasive are, I guess, you want to call it, our beliefs are and how people think similarly? And it was pretty interesting because we've created, I think, a common link and a common understanding across the world on how to approach things independent of language, business.

So I do think even if it isn't explicitly going through a particular conglomerate and how they approach things, just the general lessons of what we've learned at Constellation over the last few decades tend to be everywhere, which is surprising to me. It was surprising to me how far it was.

In a decentralized organization, it's always fascinating to see that, right? So I was pretty happy to realize that Larry came to those conclusions.

Interesting. Secondly, it's been a pretty long [indiscernible].

Mark Leonard initially floated the idea of perhaps one day Constellation having to look outside of the world of vertical market software to deploy capital. Given the level of capital you've deployed continuing in software, has that effort taken more of a backseat?

Or is it still ongoing? Or how is the level or desire evolved over with that potential initiative?

I think we're -- obviously, we're mostly interested right now in -- with our ability to get more capital out, we're interested in focusing on software and making sure we're doing software as best we possibly can. So it isn't -- it's something we discuss occasionally, but it's not really the focus right now.

And there's so much more for us to do in the software world. So I wish I could say we're all done, we're finished, but there's too much stuff to do.

We're constantly learning. And our teams -- like I think our teams have been -- are able to do larger acquisitions now, which has opened up a whole new round for us.

And we couldn't have done that 15, 20 years ago. So just keeps...

Yes. Good.

With the large -- and that's an interesting point because I think a good part or one way people are attracted and passionate to continue working at Constellation is just, it seems like a robust amount of opportunities for career development and learning even different [ fields ]. You mean is that something that people are really happy about and [ talk to you about ]?

For sure. I did something Volaris [indiscernible] 3, 4 years ago at a big event where we had like 1,200 people in London.

And I interviewed 5 people on stage and 3 of the 5 were interns overseeing portfolios. But that was 15 years later after working for us for 15 years.

So there is lots of carrier opportunities at Constellation and it needs to be. And in a decentralized environment, you create a lot more career opportunities.

When you're in a large functional organization, you hit career sort of growth opportunities, walls, if you want to call it, aren't as possible here. Many more people in our world can run a business inside of constellation.

They can take another role. So I think it's kind of interesting place to work, and we're super keen on continuing to develop our people across the world.

It's the single most important thing we do.

No, just thank everybody for dialing in and asking a bunch of good questions, and we're looking forward to getting through this quarter and chatting with you again in 3 months. So over and out from Toronto, and have a great rest of the day and rest of the summer.

Bye.

Thank you.