Thanks, Michelle. Good morning.

Welcome to Comfort Systems USA's third quarter 2026 earnings call. Our comments today, as well as our press releases, contain forward-looking statements within the meaning of the applicable securities laws and regulations.

What we will say today is based upon the current plans and expectations of Comfort Systems USA. Those plans and expectations include risks and uncertainties that might cause actual future activities and results of our operations to be materially different from those set forth in our comments.

You can read a detailed listing and commentary concerning our specific risk factors in our most recent Form 10-K and Form 10-Q, as well as in our press release covering these earnings. A slide presentation is provided as a companion to our remarks and is posted on the Investor Relations section of the company's website found at comfortsystemsusa.com.

Joining me on the call today are Brian Lane, Chief Executive Officer, Trent McKenna, President, and Bill George, Chief Financial Officer. Brian will open our remarks.

All right. Thanks, Julie.

Good morning, and thank you for joining us for the call today. Our amazing teams across the country continue to deliver excellent results for our customers.

They have delivered results that exceed even our recent record results. We earned $13.91 per share this quarter, which is $5.66 more than we earned in the same quarter last year.

Our mechanical business had an extraordinary increase in activity levels and profitability. Our electrical segment once again achieved spectacular organic revenue growth at very high margins.

Construction is driving most of our results, but service revenue and profit also grew by double-digit percentages. Our bookings were strong, and our backlog at the end of the quarter grew to a new high of $17.8 billion.

As a result of continuing strong demand, we achieved an eighth consecutive sequential backlog increase. We continue to book work with good margins and good working conditions for our valuable people.

We enter the fourth quarter of 2026 with $8.4 billion more in backlog than last year at this time. I am happy to announce the acquisition of two companies on August 1st: Boyles Electrical located in Melbourne, Florida, and NeoPod, a modular company for the healthcare industry based in New Braunfels, Texas.

We are thrilled to have these two companies join the Comfort Systems USA family of companies, and expect that they will contribute over $100 million in annual revenue at good margins. Today, we increased our quarterly dividend by 10 cents to a dollar per share, and we actively purchased shares during the quarter.

With solid bookings and great demand, we expect continuing growth and strong results for the remainder of 2026 and 2027. In a moment, Trent and Bill will discuss our results and outlook, but I want to first discuss another announcement that we made yesterday.

Last night, we announced Bill's decision to retire from the company after 30 years of amazing service. Bill helped found the company in 1997, and has led the financial strategy of the company since 2005.

As importantly, since 2008, Bill has led the extraordinarily successful acquisition efforts of the company. Since 2011, Bill has also worked with many of our companies in an operational leadership role.

I am deeply grateful to Bill for his steady leadership, thoughtful advice, and the integrity that he has brought to Comfort Systems USA every day. He has been a great partner to me.

I, along with the rest of the leadership team, wish Bill the absolute best in his well-deserved retirement. Bill will be transitioning from the CFO role shortly after we report our 2026 results, and then will continue with the company for a period to transition his other roles.

We also announced that Julie will be promoted to CFO as Bill's successor, and I also want to congratulate her. Julie has been an integral member of our finance and accounting team for many years, and I have the utmost confidence that she is the perfect person to lead the organization through Bill's transition.

Before Bill discusses our results, I've asked him to also give his thoughts about the upcoming changes. Bill?

Thanks, Brian. It's a poignant day for me, and I do want to take just a few minutes and explain why I am utterly confident that Comfort Systems USA will continue to have the best management team in our industry long after I'm gone.

Over 30 years, I've done a lot of things at Comfort: legal, financial, acquisitions, operations. The main three things I do today are financial, operational, and acquisitions.

The great news is that in the financial area, our depth is extraordinary. Frankly, it's amazing.

Right down the line, starting with Julie, we are stacked with unbelievable talent in our financial area. Honestly, the day after I leave this role, Comfort Systems will still have the best CFO in construction and the best finance department in construction.

You guys know Julie. I've been hearing from people for the last 24 hours, both internal and external, how happy and relieved they are that Julie will be our next CFO.

Operations is the part that I'll miss the most, but the good news is that our business isn't run in Houston. It's not even close to being run in Houston, it's run in the field.

As long as we maintain our culture and stay out of the way of the great companies we have, we'll always be fine. I want to spend just a little bit more time commenting on acquisitions.

Every acquisition we've ever done at Comfort Systems USA was done by a team of people. Every company we've ever bought, we pick companies, we certainly qualify companies, but we don't pick companies, they pick us.

We give them unlimited time to do and unlimited access to do due diligence on us. The reason that we continually win in the acquisition area, frequently buying companies at lower prices than other people are offering them, is because we have worked so hard to make Comfort Systems the best place on planet Earth for somebody who has made their life's work building a great contractor, a home for their legacy.

If we continue doing that, we'll continue winning, because that's why we win in acquisitions. After I leave, we'll continue to have a great team.

We have a Vice President of Acquisitions who was carefully selected as a person with the right judgment and temperament to sit in the middle of the team and facilitate things. He's somebody who won't get in the way of Comfort's message, which is all I've ever tried to do.

The people who do our acquisitions include Brian Lane, they include our CEO, our President, our Chief Operating Officer, our regional leaders. Our head of innovation is very important in acquisitions, guy named Briston Blair.

Most important, the thing that makes us win in acquisitions is our field leadership. They're always involved in the acquisitions that we do.

When we buy companies where there's overlap, we face that and solve that before we buy a company. We don't have mysteries and we'll figure things out later, and I think we'll continue to do that.

I think Comfort Systems acquisitions is on a great roll. I think that will continue.

I think we'll continue to be a trustworthy, disciplined company that does acquisitions with conviction. Above all, we'll be patient.

I would say the key to our acquisition program is that we refuse to be hurried. We're fast, actually, and when we're ready to go, we go, but we never have to hurry.

I'm very, very confident about the future of acquisitions at Comfort Systems. I just want to say one more thing about Comfort, which is how great Comfort is and why I believe it will stay great, and that's our people and our culture.

This company is filled from top to bottom with amazing people. I don't think there's anybody who understands that as well as I do, and there's a lot of people who really understand it.

In Houston, every one of our crucial functions is led by a great leader: tax, internal audit, risk management, treasury, insurance, every one of them. Most people on this call will never meet the great people we have, people with names like Adam and Amy, and Anna, and Barbara, and Byron, and Elvira, and George, Ela, J., Jordan, Julie, Judy, Michael, Wayne, and more.

I couldn't possibly name them all. The unbelievable group of people with very long tenures at the company, very, very dedicated to our company, and very, very dedicated to its culture.

Then we have perhaps even surpassing that at our subsidiaries, at our companies in all of these states: Texas, North Carolina, New York, Florida, South Carolina, Kentucky. Did I mention Texas?

We have unbelievable groups of people who have pulled together to create something that's unmatched in their markets. That culture has led to the staying power that we have in all our markets.

It's led to us making money every year of our existence, to us having free cash flow every year of our existence, to us being the best positioned when markets became great to really take advantage of it. It's that, everything I've just described, is what created the great results that I'm going to review right now.

Revenue for the third quarter of 2026 was $3.8 billion, an increase of $1.3 billion or 54% compared to last year. Electrical segment revenue grew by 85%, and mechanical revenue grew by 43%.

Through 9 months, our same-store revenue increased 47%, and we expect full-year same-store revenue will also be in the mid-40% range. We estimate that our same-store full-year 2027 revenue will grow in the mid-20% range.

Gross profit was $934 million for the third quarter of 2026, $327 million higher than 1 year ago. Our gross profit percentage remained consistent at a strong 24.8% this quarter.

Quarterly gross profit percentage in our mechanical segment increased to 25.1% this quarter compared to 24.3% last year. Despite electrical segment revenue growth of over 80%, margins in our electrical segment slipped to 24.3% this quarter as compared to 26.2% in the third quarter of 2025.

Last year, we noted that our electrical results were positively impacted in the third quarter by discrete gains, including when a customer emerged from bankruptcy. Those gains improved electrical segment margins last year by approximately 1.8%, and without that effect, electrical margins were roughly flat.

We currently expect that 2027 gross profit margins are likely to continue in the strong ranges that we've achieved in average over recent quarters. SG&A expense for the quarter was $314 million or 8.3% of revenue, compared to $230 million or 9.4% of revenue in the third quarter of 2025, and we continue to invest to support our extraordinary growth.

Our operating income increased by 65% from last year, from $379 million in the third quarter of 2025 to $625 million for the third quarter of 2026. Our operating income percentage increased to 16.6% this quarter from 15.5% in the prior year.

Our year-to-date tax rate was 23%, and we expect our tax rate to continue to be around 23% for the rest of 2026 and in 2027. After considering all these factors, net income for the third quarter of 2026 was $489 million or $13.91 per share, as compared to net income for the third quarter of 2025 of $292 million or $8.25 per share.

Thanks to great execution by our people, EBITDA increased by 61% to $668 million this quarter from a strong $414 million in the third quarter of 2025. Our trailing 12-month EBITDA is now approximately $2.3 billion.

Free cash flow for the third quarter of 2026 was $806 million, and year-to-date, our cash flow is roughly $2 billion. In addition to strong earnings, we benefited from advantageous payment terms and favorable collections from large technology customers.

Our cash flow this quarter was also higher due to $166 million in income tax payments that we were able to defer until the fourth quarter of this year. In addition to the modular expansions that are underway, we invested over $100 million in capex to expand prefabrication facilities for our traditional construction business.

We estimate that full-year capex will be 4 to 5% of revenue, and we're likely to continue to incur capex at elevated levels during the first half of 2027. We actively purchased our shares this quarter, and year-to-date, we've returned $349 million to our shareholders and we've retired approximately 212,000 shares.

Once again, expressing my gratitude to the people I mentioned earlier, and to the people on this call who've supported us for so long, I'm going to turn the time over to Trent. Trent?

Thanks, Bill. Before I start, I just want to thank you for the time I've been able to work with you over the last 22 years, and for taking the chance and hiring me back 22 years ago.

I appreciate it, Bill. I'm going to discuss our operations and outlook.

Our backlog at the end of the third quarter jumped to a record $17.8 billion. Since last year at this time, our backlog has increased by $8.4 billion or 90%.

On a sequential basis, backlog increased by $3.8 billion or 27%. Third quarter bookings were especially strong in the technology sector, both in our traditional construction business as well as the modular part of our business.

Industrial customers accounted for 76% of total revenue in the first 9 months of 2026, and they are major drivers of pipeline and backlog. Technology, which is included in industrial, was 59% of our revenue, a strong increase from 42% in the prior year.

Our modular operations continue to grow with our established customers, and we are building products for new customers, including a frontier lab and co-lo providers. Although we have not yet entered new arrangements that we would consider programmatic, we continue to explore deeper partnerships with these and other potential strategic partners.

Institutional markets, which include education, healthcare, and government, remain strong and represent 16% of our revenue, and the commercial sector provided about 8% of revenue. Most of our service revenue is for commercial customers.

Overall, our companies continue to execute at high margins. However, this quarter, we experienced a higher proportion of equipment and materials in our cost mix that augmented our sequential revenue growth, but dampened sequential margins.

Construction accounted for 90% of our revenue, with projects for new buildings representing 75% and existing building construction 15%. We include modular in new building construction, and year-to-date, modular was 17% of our revenue.

We remain on track to have over 4 million square feet of space in our modular businesses by the end of 2026, and with ongoing orders and investments to meet that demand, we expect to have approximately 6 million square feet of capacity by the end of 2027. We also continue to evaluate further investments based on continuing strong demand and customer needs.

Service revenue was higher by 13% this quarter, but with faster growth in construction, service represents 10% of total revenue. Service profitability was stronger than ever this quarter, and service remains a growing and reliable source of profit and cash flow.

I cannot say enough about the amazing team of craft professionals that we have working hard for our customers every single day. Thanks to the teams that are working across the country, we are optimistic about our future.

I want to close by joining Brian and Bill in thanking our over 25,000 employees for their hard work and dedication. I will now turn it back over to Michelle for questions.

Thank you.

Hey, good morning, guys. Congrats on another great quarter.

Julie, congrats on your promotion.

Thank you, Adam.

Bill, congrats to you on your retirement. You'll be missed, Bill.

Hey, Trent, I wanted to start with you on the sequential margin decline. Can you give a little bit more color on that and how should we think about that as we look ahead to margins in the next few quarters?

I'll jump in and then I'm sure Bill will want to add a little bit to it. Sequentially, we carried a higher proportion of equipment and materials in our cost mix this quarter.

That's going to add revenue, but equipment doesn't carry the margin our labor does. The percentage comes down even as the dollars go up.

Anyway, the way I think about it, I'd be careful about focusing on one quarter. Some of this is a little bit lumpy depending on what the mix is, and that's what we experienced in the prior quarter.

Yeah. There's other things that contributed to the slight decline.

One of the things that happens is after we do acquisitions, we have purchase adjustments, and certain of those purchase adjustments relate to the changes in the WIP schedule for jobs that were ongoing at the time we closed. Because of that, those purchase adjustments roll through our costs.

Those adjustments were like $12 million higher in the third quarter than they were in the second quarter. That was a straight hit to margins.

We also have one large company in electrical, it's over half of our electrical segment. That company had gross margins in the 30s in the second quarter, and their gross margins were still unfathomably high in the third quarter, and above average for our company, but they weren't in the 30s.

There's discrete things that happened, I think the business has never been better. The other thing is, when you see this kind of a jump in revenue, like 80% organic revenue growth in the electrical, you know a couple of things.

One, you know you have a big denominator for your fraction, but you also know that we're early in a lot of jobs, and we don't pull out high margins early in jobs, right? We wait and we keep contingency, we wait and see how they go.

I don't think there's anything but good news in this quarter, Adam.

Then, Bill, just to pull on that thread a little bit, when you say we're early in jobs, when should we think about the portfolio maturing a little bit?

We've reached a point today where it's really hard to answer that question because some of these jobs are multi-year. I will say you begin to see profit changes when you're 40, 60, 70% complete.

These big jobs, the way that you recognize revenue in our industry is you estimate the remaining cost it's going to take you to finish a contractual obligation. Early in a job, you don't know what's going to happen.

You don't know how much it's going to rain. You don't know what the people who are working next to you as other subcontractors are going to do.

Companies that stay in business, they keep money in the job for the things they don't know. As you get deeper and deeper into those jobs, you begin to get a better view.

For us historically, we've had net gain on jobs every year of our existence and they've been bigger than ever lately. I would say this quarter definitely early in jobs.

I'd say next quarter will be on average early in jobs, although there's always closeouts in fourth quarters to some extent just because of the nature of year's ending and what customers want you to do. I guess it's hard to say.

I think we'll get back to average first half of next year.

Okay. Then, obviously, unbelievable awards in the quarter.

In terms of what's driving that, would you say that's your traditional data center customers, or are there some new customers starting to creep in?

They're definitely new customers creeping in, and demand is just very strong. I think part of it's our customers are committing earlier and for larger scopes because they know that capacity like ours is scarce right now.

We continue to be really selective in what we're looking at. It's kind of all of the above.

It's a market right now where the governor isn't demand, the governor's really just labor, the ability to be able to get the labor to the work, and our customers realize that, they're wanting to lock up our capacity. That's why you see what the backlog's doing.

Adam, it's a real tribute to the folks out there working and the quality of work they're doing that we're getting this much repeat business from very sophisticated customers. Really tip your hat to the folks and what they're producing.

Thanks, guys.

All right. See ya.

Hi, good morning. You talked about building some modular products for new customers, including frontier lab and co-lo providers.

Can you just clarify if the 6 million square feet of capacity you expect by 2027 is the incremental 1 million square feet versus what you targeted by mid-year, is that related to frontier labs or new co-los, or is that primarily your two hyperscaler customers? Then can you just expand on those conversations with the frontier lab and co-los?

What are they asking from you in terms of scope of work, potential output?

The build-out that we just talked about to 6 million is based on our current programmatic build-outs with the customers that we already have. On the new customer front, yes, we're building product, right, for frontier lab and co-location providers, and that's real work, and we're proud of it.

We haven't yet entered arrangements with them that we would call programmatic, and we're not going to call them that until we believe they are programmatic, right? The build-out that we've announced is all for existing capacity that is with our current customer base in modular.

If you're living in the paradigm of the last 30 years like I sometimes do, you might say, "Oh, they haven't yet sold business to these people." The reality is we haven't yet reached terms where we're willing to accept business from these people.

Right now, one of the things that's front and center for us is the ability to pay, right? Typically our programmatic people have made 5-year volume commitments to us, and then we have gone out and spent large amounts of capex to build space that's going to primarily build their product.

I think not just in modular, but across our business, we have a long history and a long set of experience that really, really takes into account the financial strength of our counterparties. Not that these people don't have financial strength because they do, but we think there are various considerations as you compare the customers that we do work for.

Yeah. One other thing, too, and I'm sure I'm not telling you anything you haven't already realized, but modular's 17% of our revenue.

It's growing, but staying 17% because the rest of our business is also seeing really great demand as well.

Got it. All super clear.

Then, Bill, you referenced the projects in sort of earlier stages of completion. If you look at catch-up adjustments, the impact on revenues this quarter, I think it stepped down to 4.5 from closer to 8% of revenues in the first half of the year.

Am I thinking about this right? Because if you kind of strip those impacts out, it looks like underlying gross margins actually continued to increase sequentially.

Is that the right way to think about core execution, and does that give you confidence baseline margins can continue to expand as new projects mature?

I was with you all the way to the last sentence, because we do not represent that margins will go up in the future. We do know that things that have been driving them up are still apparent for us and, if anything, they're at least as strong as ever.

I will say what you said plus things I can see that nobody could see unless they were sitting in Julie's or eyes or Michael's seat, such as this point about the mix in our costs that Trent mentioned of a higher part of our cost by a couple percentage points, our cost of goods sold has more material and equipment passing through it. There's nobody who runs a business for us who worries about their return on wire.

They worry about their return on electricians and plumbers. Our gross profit per hour worked continues to get better.

I agree, and more.

Great. Thanks so much, and congrats, Bill and Julie.

Thanks.

Great. Thank you.

Good morning. Congratulations, Bill, and congratulations, Julie.

I just want to ask on the mechanical backlog, which saw a more meaningful jump than the electrical backlog, and was wondering if you could break down the makeup of that within maybe modular and the traditional business?

Backlog's really lumpy. I don't think there's any different trend between the two.

I just think in any given quarter, one might book more, one might book less. One might book vastly more and the other might book even more vastly more the way things are going.

At some point, I don't think there's any differential there. I think all of any of our 20 of our 40 of let's say we have 40 businesses, 20 are highly exposed to the super cycle.

Those people have more work than they can do. They're choosing how much they take.

It's really astonishing that our backlog went up this much in summer months. Think about the burn.

Look at our revenue. That's a really, really crazy amount of bookings.

I will say what's also crazy about it is that Brian Lane approved it. Brian Lane, Trent, get on calls for the biggest jobs and they talk through what are we doing to make sure we're safe.

We have a whole program for large projects where we monitor them even more closely. By the way, we're not kidding ourselves that anybody in Houston is performing anything, but we do really try to stay on top of it.

I don't think our backlog's the best it's ever been, Brian.

Best I've ever seen it, Bill, and the activity level out there is as strong as I've ever seen it.

All right. Then can I ask a follow-up on backlog?

Given that you're booking, it seems, larger projects, how should we think about the duration of that backlog? I know historically it's been under a year, but can you tell us how that has changed in recent quarters?

I think if you look at the backlog of what we're in now, we're obviously winning larger projects. I think the timeframe is probably pushed out to 18 to 24 months, as opposed to 10 years ago, 9 months was the best look-see we had, for sure.

The size of these jobs, the duration of them, where they are, has extended the backlog to roughly 18 to 24 months. I think, Sangeeta, would be a good framework.

Got it. Thank you, Brian.

When and

Go ahead.

I was just going to say, Sangeeta, one way to look at it is just that when I look at the backlog right now and what it did in the quarter, right, is demand for what our people do remains well in excess of what the industry can actually deliver. We're being asked to do and deliver more of it than ever before.

That's what I really take away from how strong that backlog was in this quarter.

Got it. Thank you, Trent.

That's very textured. You have the ultimate customer, but outside of modular, and I know this came up in the context of modular, but outside of modular, the payment obligation is actually typically coming from a large national GC, right?

Then in modular, we do get mobilization payments and then we have payments upon delivery. Typically at this point we're vastly ahead with our programmatic customers, and based on very open with each other negotiations.

The other thing that you look at is what is the labor that you're doing, whatever you're building, are you building something that other people would want, right? Because we have lien rights in our business, they go back to English common law, right, 1,000 years ago, where we have the right to seize the building if we don't get paid.

One of the things that we've learned to look at over the years is, is this a building that there would be many, many people who would want, or is this a building like we're involved in a wafer construction, a construction of a plant that made wafer using a certain technology where we knew that that large and very expensive building would be pretty useless to anybody but the person building it? We, in that case, asked for very special terms, where we were paid every 2 weeks and we had the right to leave on very short notice.

It's varied. There are many, many things that you take into account when you try to figure this out.

By the way, there are people rolling into modular now who are taking large books of business who we think are agreeing to payment terms where they are banking some people. If we lose work on that basis, we don't mind it.

): Okay. Maybe another sort of point of concern out there, just with the technology or data center project opportunities that are out there that you're looking at that aren't reflected in huge backlog today, presumably over the next few years, I guess any sense for how many of those you're looking at already have power connections figured out?

Pretty much the things we're looking at as prospects, the power connection is usually already solved.

Right. If anybody stroked a check for us to do design work, to make our engineers

Because of where we sit in the cycle, right, we're pretty late-cycle player in the design of a building. For our traditional construction business, I'm not aware of a lot of things that we're looking at prospect pipeline-wise where the power hasn't already been solved.

I got to tell you, it really hasn't delayed anything we're seeing yet, Brent. A lot of these data centers is generating their own power anyway, these folks are going to solve their problem one way or another.

): Okay. Just a quick technical question.

The operating margins in the mechanical segment come up a lot this year, including the third quarter. Is that primarily due to better profit margins in modular compared to last year, at least on an operating basis?

Essentially, are you getting better fixed-cost leverage there, and that's the big

In the mix, that's the single largest driver in the mix. There is actually higher margins for the 9 months for sure in pretty much all parts of that work.

Modular has really closed the gap with the rest of the business to come really up to parity with our average margins. Of course, mathematically, that changes the mix.

I'm just going to throw props to one company that's doing an unbelievable job on piping, really throughout the country, and their margins have improved significantly and is doing a heck of a job.

Yep. ): Very good.

Okay. Thanks, all.

Appreciate it.

Hi, good morning, and congrats, Bill and Julie, both truly very well-deserved. Congrats.

My first question, could you talk about the magnitude of the orders that you took in the quarter? It's sort of billions of dollars more than taking in any other quarter.

Maybe I know that the demand is obviously really strong, but just kind of what drove that decision, the confidence around being able to phase it out and things of that nature?

Josh, the first part of that is that the timing of the work, we can handle, we have the personnel to do it, the locations are working in our favor. The most important thing is labor availability of who and when it's got to go.

As Bill said earlier, when we work with our operating companies looking at these opportunities, where they are, we're very confident we can handle all the work we have right now with the forces we have. We're also very fortunate, the company of this size, that the companies collaborate, work together.

Add that on to have a good traveling workforce, when you add all that up, we're very confident executing the level of backlog we have today.

No, that's great to hear and obviously great backlog. I guess my follow-up, could I ask about the margin comment about the mix of materials?

Because obviously understand the mathematical reason for that, but over the years, it's very rare for you to call out that mix as a driver. I was wondering if anything large specifically happened this quarter where that mix was more impactful than usual.

Material and equipment was about 2% higher than it had averaged in the first 6 months. The only other time we've brought that up was during COVID because there was so much inflation in materials and equipment.

I think that the dollars of them just became such a bigger part of our cost mix. I think in this case, a big part of it is the earliness of the jobs.

A lot of stuff gets delivered. The earlier in a job, you're doing very, very long pipe runs and things like that.

I think some of that actually comes back to the earliness of the job. Some of it's just lumpy, and that's just the way it is.

To Trent, what do you

Everything you said, Bill, is correct. It's really where the large projects are in their life cycle and where we are in ordering that equipment.

Great. Thanks for the color and congrats on the quarter.

Thanks.

Thanks for taking my questions. Good morning.

The first question I want to ask about is on the backlog. Obviously very strong result, but we don't have insight into the individual conversations that you're having.

I am wondering with all the political rhetoric around data center moratoriums heading into the election, other regulations that governors are proposing around data center construction, just curious if you've seen any change in any project schedules or push out in the pipeline in any of the states.

I would say, and having talked to one of our biggest companies in the last day or two, the answer is generally no. I don't know of a project where one of these considerations has changed its trajectory.

I will say, I think on a more macro level, I have a sense, and I've talked with people who've given me this sense, this may push more of the work nationally into more sort of friendly jurisdictions. If you were a large hyperscaler and you were laying your bets on which states you wanted to do work in, I think where you place your bets, where you go do your work and make your commitments is going to be influenced by your perception of whether the governor in that state wants people in that state to have jobs, right?

I'm not saying it has no effect, but I don't think at a micro level we've seen any effect whatsoever.

Tim, I can tell you, for the work we have today, what we're seeing here at Comfort Systems, we have not been delayed by any moratorium.

One thing more to think about, too, is that modular's even less exposed, because it ships to wherever the customer's site is. It's also something to keep in mind for that 17% of our business.

That's really helpful color. Thank you.

Just on the modular business, by the way, just my last question here. I'm curious why you haven't added any more programmatic customers in your modular business.

I know you touched on this earlier, but the demand is clearly there. Could you just walk me through that decision matrix around that in a little bit more?

Thank you.

I'll I'll start. I'll start.

This happens in the real world, right? We're talking about 250,000, 600,000, 700,000 square foot buildings that are being built out with rows of robots and automated paint booths and turntables, and have to be staffed with, thousands ultimately of people.

We think we've grown unbelievably fast. We're not This is not scalable the way you can just print another copy of software.

Honestly, we have tended to stick with our two big customers who've kept us, kept us happy together for, in one case, for 20 years, because we know them, it works, they want more than we can buy. They're they can't buy enough from us to the point where they're looking at other sources of supply.

Another a another way of ask the other side of that question is, why would we divert from a sure thing to sort of a mystery, right?

Yeah. Yep.

Makes total sense. Thanks for the color, guys.

All right. Take care.

Thanks. Hey, good morning.

Again, Bill, congrats on your retirement, and Julie, congrats on the CFO role, well-deserved. Hey, Trent, you mentioned earlier that your customers understand that labor is the governor.

I guess if that's the case, what lever would expand that fastest? Would it be training, would it be modular automation or digital field tools?

What kind of growth in your headcount does mid-20s growth in '27 require?

It's going to require growth. We're over 25,000 today, right?

We've grown that kind of the way we've always done it, right, where our companies strive to be the best place for craft professionals to work. We pay well, we train relentlessly, and we are always trying to keep people busy on good projects with good conditions.

That's what craft professional is interested in. Obviously, our traveling workforce has helped a lot in this, and our contracted craft operations at Kodiak and Pivot has helped as well, and acquisitions like Hunt and Boyles have already let us put people where the work is, right?

There's a lot of ways we're doing it. The one key thing about us is that if we've taken work, it means we've planned against every dollar in that backlog.

That's one thing about us. When we look at something, to Bill's point earlier when we talked about Trent and Brian looking at work, the question's always, do we have the labor to do it?

Then it's labor planned before we ever agreed to it. As we look into the future in that mid-20s, we will have to add people, but we'll just do it the way we've always done it.

That's really the easiest way to kind of sum that up is we want to be the best place for a craft professional to build their career. We'll continue to try to be that place.

There's a lot of things that go into that, but that's the easiest way to understand it.

I've been doing construction a long time, probably more than I want to remember. We really do treat people with respect and fairly here, and I think that really matters when you bring people in and they stay with us.

I think it's really important what the culture of this company is and how we get people to go work where they go work.

If it's okay, Brian, I want to add to that, because if you look at what our people, the 25,000 plus just did over the last 3 months, they were working in the hottest months of the year, many of them were on the road away from their families, and they still delivered safely on schedule with great quality for our customers. I just want to take my hat off to them and commend them on the work that they performed and how grateful I am for what they've done.

Thanks. Really helpful there.

Then maybe just asking again about what you mentioned earlier about programmatic relationships, typically involving multi-year volume commitments and enough visibility to support capex. Earlier, to another question, you stated, why would you divert from the two large programmatic customers?

I guess what would you need to see from one of the newer customers before they could kind of graduate to programmatic, and how important is the financial strength of the customer in those considerations?

I would say that the two things that you would look for are the nature of the commitment and the payment timing, right? Payment timing, and also you actually take into account many of these newer customers, they're new to building modular, hot aisle containment or modular chiller plants.

One great thing about them is they're willing to really listen to our experience and build stuff. If we build stuff that's constructible and versatile, that gives us a lot of comfort as well that it's valuable to them or to someone else, and maybe more importantly that the capacity that we set up to build for them has a lot of outlets.

None of it's mysterious, you could sit and think of it, nobody here is a genius who's thought of something new. It's exactly what you would think.

Very helpful. I'll pass it on, and congratulations again.

Thanks.

Great. Thank you, and good morning, everyone.

I just wanted to go back to the point on capacity. I guess for those operating companies that are doing larger data center work, how much capacity, to speak, is just left looking at the projects in backlog and how that carries you through next year?

I guess what I'm ultimately trying to get at is thinking about bookings the next couple quarters and whether the outsized performance this quarter maybe takes away from what we might expect looking forward.

I think that, first of all, that question doesn't have a mathematical answer. There's no gauge, right?

Your willingness to stretch to accept work has to do with work, the nature of that work, the flexibility that you're given, the timing. I would say it's certainly the case that if we book more in one quarter, we might book not as much in a future quarter.

My sense is our guys, they're going to continue to book the work, probably continue to book farther out. At this moment, the conditions are extremely supportive with no sign of that changing.

I don't know, but I can't change the fact that you don't collect a dollar twice, you don't book the same job twice, right?

Curt, we do remain disciplined on what we're going to take. The big driver for us is we're not going to take work that we can't do at a very high level.

That to me is the overriding factor when we look at these things. We have world-class people out there looking at the work, doing the work, we'll handle what we can handle and we'll see how it goes.

We'll keep booking work, I don't know, as we go farther and farther out.

Yeah. All right.

That makes sense. Then, just second, on capital deployment, given the level of cash generation, I guess I'm just curious, looking at historically as a percentage of that, what you've deployed towards M&A, how does kind of the current landscape look in terms of the ability to continue to do that, and how does that maybe impact the thinking or aggressiveness that we might expect from kind of a share repurchase perspective going forward?

We look at these decisions over years, not months or quarters. We have a very, very good acquisition pipeline, but of course pricing for acquisitions has been high.

I think that might be rolling over right now with interest rates and with the continued lack of exits for certain private people, but we'll see. Of course we've always bought our shares more aggressively when they dip, right?

Over a long period of time, we have used 15 to 20% of our free cash flow to buy shares. If you look over the time we've bought them since 2007, we buy a lot more during DeepSeek.

In 2-week in a 2-week period, we spent $200 million, right? That $200 million, by the way, more than a year ago was more to us than it is today.

We've spent $200 million this last quarter. I guess all of those things, we're trying to do exactly what you would do.

We have a lot of confidence in our future, we have a lot of confidence in stock buybacks, especially when the price is softer. We have a really good acquisition pipeline, but we will not do a deal just because we have money.

By the way, one turn of EBITDA in cash, we've had that many, many times in the past. It's a big dollar amount, but it's not a big dollar amount on the scale of what Comfort is yet.

Okay. That makes sense.

Thank you for the color.

Thanks. Good morning.

Appreciate you taking the question. Bill, Julie, congrats to you both.

Bill, quite an incredible career, and Julie, very well-deserved. I guess just one for me.

Curious if there's any areas, geographic exposure that kind of stand out this quarter, in terms of relative strength or weakness that's worth calling out. Obviously, it sounds like Texas remains quite strong, but curious if there's other regions that are notable.

Thanks.

Brian, you hit it right on the head, Texas really strong. Georgia strong and getting stronger.

Seeing a lot in Mississippi, Louisiana, we don't do much work in, but Louisiana strong. Then, of course, you've got North and South Carolina in the Southeast areas, but Texas right now is really strong.

Utah is very strong. Bill, any others?

Iowa. Oh, Iowa.

That's why I woke up. There's a new

There's a new one. Yeah.

Yeah.

If you're talking about the areas of what we consider above normal strength, those would probably be the territories.

By the way, I'm not saying we'll play there because we may choose other options, but the next phases of some of the big semiconductor work is about to hit. We have opportunities there.

I don't know that we'll take very much of that because we're so busy in Texas, but if you're just asking about geographies, there's some monster work in Arizona.

Very helpful. I appreciate it.

I'll pass it on. Congrats again.

All right. In closing, I want to thank our teams across the country for their dedication, skill, and excellence they bring to serving our customers every day.

Demand remains strong and our people continue to rise to the challenge of meeting an unprecedented need for their expertise. Conditions remain favorable for continued strong performance, and we have the resources and commitment to build on that momentum.

Thank you for your confidence, and I hope you all have a great weekend. Thank you.