Q4 2024 Comfort Systems USA Inc Earnings Call

Thomson Reuters StreetEvents
02-22

Participants

Julie Shaeff; Senior Vice President, Chief Accounting Officer; Comfort Systems USA Inc

Brian Lane; President, Chief Executive Officer, Director; Comfort Systems USA Inc

William George; Chief Financial Officer, Executive Vice President; Comfort Systems USA Inc

Brent Thielman; Analyst; D. A. Davidson & Co

Adam Thalhimer; Analyst; Thompson Davis & Co

Julio Romero; Analyst; Sidoti & Company LLC

Josh Chan; Analyst; UBS Securities LLC

Brian Brophy; Analyst; Stifel, Nicolaus & Company

Alex Dwyer; Analyst; KeyBanc Capital Markets Inc

Unidentified Participant

Presentation

Operator

Good day, and thank you for standing by. Welcome to the fourth quarter 2024 Comfort Systems USA earnings conference call. (Operator Instructions) Please be advised that today's conference is being recorded.
I would now like to hand the conference over to your first speaker today, Julie Shaeff, Chief Accounting Officer. Please go ahead.

Julie Shaeff

Thanks, Marvin. Good morning. Welcome to Comfort Systems USA's fourth quarter and full year 2024 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 in these comments. You can read a detailed listing and commentary concerning our specific risk factors in our most recent Form 10-K, 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, President and Chief Executive Officer; Trent McKenna, Chief Operating Officer; and Bill George, Chief Financial Officer.
Brian will open our remarks.

Brian Lane

All right. Thanks, Julie. Good morning, and thank you for joining our call today. Last night, we reported record annual and fourth quarter earnings and exceptional cash flow, thanks to excellent execution by our superb teams across the United States.
Same-store revenue growth for the fourth quarter was 22%, and our margins were amazing. We earned $4.09 per share this quarter, up 60% from last year, and $14.60 per share for the year compared to $9.01 in 2023. Backlog at the end of the year grew to a new all-time high of $6 billion. Our backlog is broadly based, and we continue to experience strength across virtually all sectors, including persistent strong demand from our technology customers.
Thanks to strong fourth quarter bookings, we began 2025, with same-store growth in both sequential and year-over-year backlog. 2024 operating cash flow was $849 million, laying a strong foundation for continued investment and our ongoing cash flow strength is indicative of strong underlying trends in our execution, customer relationships and prospects.
We recently acquired companies continued to surpass our expectations and we are pleased to announce that in January, we added Century Contract as our newest partner company. Century is an outstanding, well established mechanical contractor based in [Charlotte], North Carolina, and we expect they will earn about $90 million of revenue this year.
We also announced a quarterly dividend increase of $0.05 to $0.40 per share. This increase reflects our strong cash flow. And like our share repurchases, it shows our commitment to reward our shareholders. I will discuss our business and outlook in a few minutes, but first, I will turn this call over to Bill to review our financial performance. Bill?

William George

Thanks, Brian. Results are fantastic this quarter, capping a great year. Revenue for the fourth quarter of 2024 increased by 38% compared to last year, to $1.9 billion. Full year revenue for 2024 was $7 billion, an increase of 35% compared to 2023.
For the full year, our Mechanical segment revenue increased by 40%, helped by acquisitions, modular expansion and substantial organic construction and service growth. Electrical segment revenue increased by 19%. Full year same-store revenue increased by 23%, or $1.2 billion. We'll face a tough comparable in 2025 in our best estimate, is that same-store revenue will continue to rise in 2025, most likely by high single-digit percentage growth.
Gross profit was $434 million for the fourth quarter of 2024, a $154 million improvement compared to a year ago. Our gross profit percentage grew to 23.2% this quarter, compared to 20.6% for the fourth quarter of 2023. The quarterly gross profit percentage in our Electrical segment improved to 26.1% this year, compared to 22.9% last year. Margins in our Mechanical segment also increased in the quarter to 22.4%, compared to 19.8% in the fourth quarter of 2023.
Full year gross profit increased by $486 million, and our annual gross profit margin was 21% as compared to 19% in 2023. Our Electrical margin was 24.1% for 2024, while mechanical was 20.2%. As we look to 2025, we are optimistic that gross profit margins will continue to be in the strong ranges we have achieved in comparable quarters last year.
SG&A expense in the fourth quarter was $208 million, or 11.1% of revenue, compared to $160 million or 11.8% of revenue in the same quarter of 2023. For the full year, SG&A expense as a percentage of revenue was 10.4%, down from 11.0% in 2023. For the full year and on a same-store basis, SG&A was up $117 million due to ongoing investments to support our much higher activity levels.
Quarterly operating income increased from $120 million in the fourth quarter of 2023, to $226 million for the fourth quarter of 2024, an 88% increase. Thanks to the jump in gross profit margins and favorable SG&A leverage, our operating income percentage increased to 12.1% this year, from 8.9% in the prior year. For the full year, our operating income was $749 million, and we achieved a remarkable operating income percentage of 10.7%.
Our 2024 tax rate was 21.6%. We estimate that our tax rate in 2025 will be approximately 22% to 23%. After considering all these factors, net income for the fourth quarter of 2024 was $146 million, or $4.09 per share. This is a 60% improvement in quarterly earnings per share from last year. Our full year earnings per share for 2024 were $14.60, compared to $9.01 per share in the prior year, so our annual EPS is up by over 60%.
EBITDA increased by 85%, to $261 million this quarter, from $141 million in the fourth quarter of 2023. Same store EBITDA increased by over 50%. Full year 2024 EBITDA was $892 million, and EBITDA margin was 12.7%, reflecting great execution by our teams and strong demand in our markets.
Full year free cash flow for 2024 was a remarkable $744 million. We continue to benefit from advanced payments and operating cash flow, again far exceeded our earnings by $327 million on a trailing 12-month basis. So we are again well ahead of earnings in collecting our cash.
In addition, our cash flow for the second half of 2024 was lifted by approximately $80 million of tax payments that we were allowed to defer until February of 2025 due to Hurricane [Barrel]. That catch-up payment has now been made and will reduce first quarter cash flow.
Capital expenditures in 2024 were $111 million, or a little over 1.5% of revenue, and we continue to invest in our operations and purchase vehicles to support our service business. We increased our investment in share repurchases this year and returned $58 million to shareholders in 2024 by buying over 177,000 shares at an average price of $329. Since its start, our share purchase program has retired over 10.4 million shares at an average price of $31.41, and paid our shareholders more than $320 million.
That's all I got, Brian.

Brian Lane

All right. Thanks, Bill. I am going to discuss our business and outlook.
Backlog at the end of the fourth quarter was $6 billion, a same-store increase in both sequential and year-over-year backlog. Fourth quarter bookings were strong, especially in the technology sector, and sequential backlog was up $300 million. Since last year, our backlog has increased by $800 million, or 16%. On a same-store basis, our backlog is now up by $400 million, 9% higher than at this time last year. Our overall backlog is broadly based and especially robust in our industrial sector. We are carefully selecting work that has good margins, with good working conditions for our variable workforce.
Our revenue mix continues to trend towards the industrial sector. With this sector accounting for over 60% of our volume in 2024, and continues to be a major driver of pipeline backlog. Technology, which we include in industrial, and which includes data centers and chip fab, was 33% of our revenue, a substantial increase from 21% in the prior year and advanced technology is now the largest component of our revenue.
Institutional markets including education, health care and government are also strong and represent 24% of our revenue. The commercial sector is active as well and now accounts for about 16% of revenue. Most of our service revenue is for commercial customers, so our commercial construction business is now relatively small.
Construction accounted for 84% of our revenue, with projects for new buildings representing 56%, and existing building construction 28%. Project pipelines remain at unprecedented levels. We include modular and new building construction, and year-to-date modular was 17% of our revenue.
Service revenue was up 8% this year on an absolute basis. But with faster growth in construction, service is now 16% of total revenue. Our overall service business achieved a record $1.1 billion in revenue for 2024 and service continues to be a growing and reliable source of profit and cash flow.
With record broad-based backlog, healthy pipelines, persistent demand in advanced technology, onshoring and especially our unmatched workforce. We expect continuing strong results in 2025.
I want to close by thanking our over 18,000 employees for their hard work and dedication. Our success is a direct result of the people that serve our customers every single day. I will now turn it back over to Marvin for questions. Thank you.

Question and Answer Session

Operator

Thank you (Operator Instructions)
Brent Thielman, D.A. Davidson.

Brent Thielman

Thanks. Good morning. Congrats on a great finish of the year. I guess, Brian or Bill, I mean, I think your comments on your ability to kind of sustain these elevated margin levels in 2025 is pretty notable. I'm just wondering, is that a function of the terms that you continue to see come in on new awards? Is it the schedules you see laid out for 2025 across the different territories?
Just wanting to understand gives you the confidence around that, just given such strong margins by historical standards.

Brian Lane

Yeah, for sure, Brent. So as you know, Brent, you've been around this business a long time. It's a host of things, and it starts with good private selection, working for good customers, leading on to -- doing a great job estimating and evaluating the risk of the projects that we're looking at. And right now, there's just a lot of work out there that's in our wheelhouse, in our sweet spot. So we're doing a really good job of executing it.
But it all comes down to the folks in the field that are actually doing the work with tools in their hands. We are very fortunate to have the workforce we do and the care they take with our customers. So it's a combination of a lot of things are going to drive margins and we have to do all of them well, which we're doing right now.

Brent Thielman

Yeah, okay. And then on modular, I mean just taking the year-to-date 17% of revenue, I mean it looks like the business is up roughly 50% in 2024.
Could you just talk about your ability to continue to see growth in modular in '25? I know there's limitations in terms of the capacity, but I just wanted to get a handle around how you're thinking of that in the 2025.

William George

Right. So Brent, we had a big bang increase in '23 based on some giant orders and reassurances that we got in 2022. We continue to grow that, but now when we add space, we'll add 200,000 square feet, and we will put a lot more focus on productivity, automation.
So we really see that continuing to grow just gradually, and I just -- our guys are so good at really providing a great product to the customer and really helping not just with -- you don't just have to build it, you have to help them design it. And I think we're just excellent at that, and really, we stand in all of our guys. But I think you'll just continue to see sort of incremental growth.

Brent Thielman

Okay. And Bill, I mean, with the focus on productivity automation, presumably that could put some upward pressure on margins in that particular area. Is that fair?

William George

I think it's supportive of margins. These guys are getting really good margin. And they have big sophisticated customers. So I wouldn't necessarily predict.
I think that, that space will grow faster than Comfort as a whole. So it will be supportive of those growth projections -- same-store growth projections we just gave. But I think our margin guidance for them would be the same as our margin guidance for the rest of the business, which is overall, we can't believe the margins that they achieved in '24, but we really see every -- we have every reason to think they'll do it again in '25.

Brent Thielman

Got it. Last one, if I could. Bill, I mean, the CapEx stepped up here in the fourth quarter, maybe it's associated with modular or something else. Just -- what that was related to? And thoughts on CapEx in 2025?

William George

So dollars of CapEx, they're up, they're up. If you look fourth quarter to fourth quarter, they're still down as a percent of revenue. They're up sequentially, but that's almost always going to happen from the big volumes of the third quarter. So as a percentage, they're up sequentially.
I will say our first use of every dollar we get, our very first use is to reinvest it in the business where it can help us be a better company and serve our customers better. We also occasionally, nowadays, will buy the buildings that are -- that we run our businesses from opportunistically just because they're often in a medium-sized town and they're pretty special buildings. And we need -- we're really investing in those buildings and frequently the most logical way to do that is to simply buy the building from the landlord.
So sometimes you'll see -- I think we'll probably buy a couple of our buildings this year. We're in some conversations that will probably lead to that. And so I think our backlog -- I mean, I think our CapEx will stay at about the same percentage next year as this year.

Operator

Adam Thalhimer, Thompson Davis.

Adam Thalhimer

Hey. Good morning, guys. Congrats on a great Q4. I wanted to start on the persistent strong demand from technology customers. Can you give us a little more color on that? And maybe just some anecdotes what you're seeing from the current bidding environment?

Brian Lane

Yeah. So I'll go first and then Bill can add on to it. But probably a lot of concern a couple of months ago would have released. But right now, we're seeing heavy activity, particularly on the data center front. There's been no let up on demand and what they want us to build. In fact, they've let us know that they want to make sure that we're still all in on building in which we are. So we see no let up at all, Adam, on that front.

William George

One of the things people forget is that the amounts that have been announced by the hyperscalers in particular, are vastly higher than anything, any year that's ever happened in the past and probably not possible to start with. So even if there's a pullback, it's a pull back to a lower amount that's still impossible in the time frames that they're talking about. And it's definitely the case with the people we build data centers for that if they're going to slow down, they don't know it yet.

Adam Thalhimer

Helpful. And then to what extent are you guys getting? Or do you have visibility, would you say into 2026?

William George

Well, compared to any other time you would have asked that question more than usual, right?

Brian Lane

It's actually -- we have more book backlog in the 2026 than we've ever had at Comfort Systems.

William George

2026 is still a year away.

Brian Lane

Yes, still a ways away.

William George

At some point, no, I mean we've never had as much visibility as we've had. We've never been as fully sold for the coming year and into the next year as we are now. And really, the project sizes are very big. So they can frequently be multiyear.
The one thing that happens is the starts are lumpy, right. And you've been -- you followed us for so long. You know that. But as of right now, frequently, when a start is delayed, it's like a relief for our subsidiary because they've got so much going on that they're starting to kind of count on some of those starts being delayed.

Brian Lane

But you know Adam, it's also great for our workforce that they see that this work coming into 2026 and see how much work that is out there, because we have a lot of work to do this year. So it's a really good sign for the folks in the field, too.

Adam Thalhimer

Nice. And then just quickly, Bill, how big is that Q1 tax payment?

William George

So we paid $80 million that should have been paid last year. So if your ZIP code was in the federally declared disaster area, they basically gave you relief so that you didn't need to make your quarterly estimated tax payments for the third and fourth quarter of 2024 with no penalty or interest. And on behalf of our shareholders, we took advantage of that. And -- so we paid $80 million a week or two ago that should have been paid last year. And we'll still be making a tax payment this quarter.
Our normal estimated tax payment for our very high profits right now. So that's literally a shifting of $80 million of cash flow from this year to last year. And also in the first quarter, we have some acquisition-related payments, where these acquisitions are doing so well that when we pay out some of their earn-outs and [staff]. Once you're -- the payouts that you make on earnouts go above the amounts that you estimated, when you accrue for the earnouts, they start to go through operating cash flow.
It is self-funding because by definition, they earned more money, but it does change the characteristics of the cash flow. So we think we'll have great cash flow this year but especially the first quarter, there is some displacement, some temporal displacement for some of the payments, all of which were really to the benefit of our shareholders, but they were all kind of at the same direction, so.

Operator

Julio Romero, Sidoti & Company.

Julio Romero

Thanks. Hey, good morning, guys. Yeah. Just thinking about the modular offering and obviously, a lot of concern about the data center demand that you're seeing. But can you maybe just give us a refresher on the other end markets that currently comprise your modular mix aside from data centers? And maybe if they don't -- if they're not served by modular now, they could be served in the future?

William George

So historically, our modular business, like when we bought it in 2011 and for the first many years, it was really focused towards pharmaceutical builds. I'd always had the data center component to it. As the data center stuff grew, essentially the data center, some really big and good customers came to us and said, we want pretty much your whole capacity. And we talked to them about the terms that we would need and the cooperation, we would need to give that to them.
So we've given -- they probably have 80%, 90%, if not more, of our capacity. We do continue to -- now we're much bigger. So we continue to reserve 10% -- a chunk of more than 10% of our capacity for the customers who grew to rely on us in other verticals. And keeping in mind that 10% -- it's 10% of a number that's grown fivefold, right? So it's still a big number.
We think this modular construction would be useful in all kinds of -- well, we know it would be, in all kinds of verticals. But it's just such a good math right now for data center and the demand for data center is so big. And really, the people who are building data centers, people think -- people kind of ask is, is this going to be the way people build data centers?
The reality is for the foreseeable future, the way that we're going to build data centers is every way it's possible to build a data center. They're really taking an all of the above approach. And I don't know. We think there's -- obviously, we're doing that much work if something we don't see or expect happens, which is data center slows down or stops, we'd have to replace that work and there would be an adjustment period.

Brian Lane

Yeah. Julio, we've done it for hospitals as well, but we're at the very beginning of module application in construction in this country. This will have a long, long runway, which is -- it's a great solution in construction, labor, et cetera, risk. It's a terrific solution.

William George

Yeah. There's amazing applications like in telecom. And there's just like a lot of places where it would be -- what's the capacity is out there, right? But today, what's done modular is like a very, very small fraction of 1% of what could be done modular.

Julio Romero

Yeah. No, agreed. Very helpful color, and we're definitely in early stages there. You said this also -- a data center is not just (inaudible) modular, but you also do some [stick] built work for data center as well. Can you maybe just talk about demand trends on that side?

Brian Lane

Yeah. It's extremely robust. You're just talking about Texas alone as one example. We're turning down -- we're actually turning down work as we're full up. There's great opportunities in Mississippi. I could go on and on, but there's plenty of opportunities for a regular bill that we're doing as well.

William George

The work we do in data center that's nonmodular is almost the same amount of revenue as we do modular. Yes. I think about it. It's a big set of opportunities for us.

Julio Romero

Really helpful. And then one more for me is just you talked about this Century Contractors acquisition you made here in the first quarter. The location seems to be pretty close to your EAS facility. Is there any kind of synergy with that facility, any modular component or benefit to Century Contractors?

William George

So they got -- that wonderful capability and complex pipe, and that's a really, really good fit for stuff we already do, but really, it's a really good fit for where -- if you were to like look at -- we have five or six businesses sort of within 100 miles of there that are doing industrial work.
If you'd have said if, I could have one more of something, a little -- if I could have more capability and one thing it would have been and what these guys do. So we're optimistic that this will be a really, really great member of that ecosystem, and they're already off to a great start.

Operator

Josh Chan, UBS.

Josh Chan

Hi. Good morning, guys. Congrats on the great quarter. I know that the gross margin is usually fairly strong in Q4, but could you just kind of help us contextualize the magnitude of strength in this Q4's gross margin? Was there anything unusual? Any closeout benefit anything like that?

William George

Brian, did you think we have good gross margins in --

Brian Lane

Yeah. It's interesting, Josh. It's a really good question because when we saw the number 2, we took -- we look twice. It comes back to -- we get really good work we're really, really getting superb execution. I can't really say enough the quality of work we're getting -- we're getting out in the field. And we're looking for work we're really good at, and it really helps in margin, so.

William George

And honestly, there really wasn't any like particular or a small handful of special closeouts that drove a bunch of that. It was really broad-based. It's not -- we've reached a point now where this -- so first of all, you have modular, those projects are built in a month or two, right? And so much the industrial goes year round that the closeout become kind of every quarter, but never more than usual. I could that could be -- here and there, there might be a quarter where there's one or two special closeouts. This was just broad-based strength and revenue running through at great margins.

Brian Lane

The other thing, Josh, our service business continues to grow. We did [$1 billion] one-off last year. If you look at the back end of (inaudible) where you get significant temperature change, got really cold in parts of the country that really accelerates your service call-out work, et cetera. So that helps our margins as well.

Josh Chan

That's a fair point. Yes. That's really encouraging. And then on the -- I guess your people-based business. So can you talk a little bit about the progress behind hiring people and the training of apprentices. How do you feel that's going relative to your demand outlook?

Brian Lane

Well, I mean, the organization is doing a terrific job, bringing folks in here, constantly recruiting. We got outstanding training programs from skilled trades all the way up to leadership training, where we got training going on every week in this organization.
The companies that we brought in here years ago. [Codiak], for example, has access to a lot of temporary labor that's really helped us manage our peaks and valleys and given us outstanding talent. So the recruiting effort's constant, will not stop. We got a lot of recruits throughout the country. And that will never stop in training. We're totally 100% committed to training. That's something that we'll never slow down on.

Josh Chan

That's great. And maybe I can ask one last one. How does the M&A pipeline look? And how do you expect that to kind of contribute through the year?

William George

So I would say the M&A pipeline is very healthy, but we are coming off of years where we did really -- last year, we did our biggest deal ever. We did two really notable deals and another great deal. So we are optimistic that we can just continue to keep doing what we've been doing in any given year. We'll do more or less based on conviction and availability. The one thing we won't ever do is rush or try to fill a quota. But right now, things are good.

Operator

Brian Brophy, Stifel.

Brian Brophy

Thanks. Good morning, everybody. Thanks for taking the question. Congrats on a very nice quarter. I wanted to ask on some of the comments around same-store sales growth for 2025. You mentioned high single-digit range. It sounds like it's a little bit of a change from some of your comments last quarter. So can you just give us a sense for what changed versus what you were mentioning?

William George

It's just math is what it is. We did not expect to roll through quite the revenue that we did in the fourth quarter. When we gave guidance on our revenue, it comes from an organic planning process, we do where we get. So the numbers, as big as ever, it's just that, that same number is not as percentage bigger -- as much percentage bigger as the bigger number that we came up with in the fourth quarter -- that made the full year bigger in the fourth quarter. So is just a --

Brian Lane

Yeah, it's not for less work. There's plenty of work out there.

William George

It's just such a big revenue number in the fourth quarter.

Brian Brophy

Understood. Yes, that's helpful. And then when I look at the implied mechanical orders, those were down a little bit versus a year ago.
Does it have to do with anything regarding the timing of modular orders? Or is there anything else that's driving that?

William George

I think you're smarter about that than I am. We -- I mean I would say that we don't -- what we are not seeing -- we're seeing -- we got really good modular orders in the fourth quarter, but we did not get the gigantic modular orders we got in particular, in December of '22, and also in December of 2023. But we had net bookings and the plants full, and it's booked full longer than it's ever been. It's a bigger plant. There doesn't seem to be a slowdown, but there's not a step change sort of increase like we had.

Brian Lane

And Brian, I know you're relatively new to us. But what's really important here is that whenever we look at work we're going to take is our manpower loading curves. Make sure we do not take too much work that we can't execute at a high level because margins, profit and cash flow are really important. So we spend a lot of time about what work we can handle and when the work is coming.

William George

There is no consideration or element in our backlog number of they're not being work to do. There's -- we double our backlog in the next 105 days if we just started extending (inaudible) to more work. We could take a lot of work we couldn't do if our goal is simply to increase our backlog.

Brian Brophy

Understood. Yes, that's really helpful. And then I guess one other one. We've been getting a lot of questions on inferencing data centers versus training data centers. What that potentially means for power, cooling needs.
Can you talk about some of the differences in terms of electrical and mechanical intensity between some of these different types of data centers? And then just broadly, when you're looking at your data center activity builds to date, how much is more cloud-based data centers versus AI data centers today?

William George

So I'll give a little bit of color about that, but I'm not really not an expert, right. There's no data center -- here's some really clear things. There's no data center we're building today that's not more dense than if we were building that same footprint 1, 3 or 5 years ago. They are bringing in way more copper, way more electricity. They're trying to put more compute power into all data centers.
The true AI data centers are barely even starting. They will probably have a smaller footprint based on just what smart people at Comfort tell me. They will have unbelievable power draws. But the black belt chips only started actually being delivered a few months ago. The first ones have gotten into servers, most of them are scheduled to be delivered in next year and the year after, right. So that is really in its infancy.
The true build for -- so everything that was on the block, everything that was being built, redesign work was done to increase the density of it. If you talk to our guys as much as they could. But I don't think we've really started, to a material extent, I don't think we've started building the true AI data centers. I think that's yet to come. And I think because of that, it's still being figured out.

Brian Brophy

Understood. And then last one for me here. Have you guys seen any changes in terms of contract terms today versus a year ago? I would assume they're essentially the same or better? And how should we think about that impacting margins and free cash flow conversion this year?

William George

So we are -- we are getting great payment terms, but we were already getting great payment terms a year and 2 years ago. And honestly, that's a constant trade-off, right? We don't really need money sooner. If you if you need -- if you're somebody that has all the money in the world and you -- there's a wide range of payment terms, we'll accept. Generally speaking, they're just very generous payment terms right now.
As far as the underlying contractual provisions, as you can imagine at a time when we're this busy, that affords us the opportunity to say no to things. So at the margin, we're doing some of that. There are GCs who have very unreasonable forms. I think that's affecting their ability to get subcontract. Actually, I know it's -- they may not know it, but I know it's affecting their ability to get subcontractors and not just with us.
But ultimately, the forms that are used in this business have been around for a long time and almost more important than the words that are in them is like there's a standard of practice. And if you do work over and over with the same people that you trust, that's the most important thing. I would rather have average terms with a great long-term partner than a better contract with people, I don't know.

Operator

Alex Dwyer, KeyBanc Capital Markets.

Alex Dwyer

Hey, guys. Good morning. Thanks for taking my questions. So I wanted to come back to the margin guidance in the flat in '25 versus '24. And just kind of walk through the biggest risks to this, and what could drive to the upside here? Whether it's like a pricing, execution, closeouts, cost inflation or -- or is there anything different about the project environment that can make it harder to execute this year and share resources?
It just seems like there's just such a great gross margin story with modular improving efficiencies and in the supply-demand imbalance. And I think there is a good amount of this acquisition amortization that rolls off. So just any thoughts on why margins couldn't actually expand this year versus last year?

William George

Right. We would never say that they couldn't. I think our hesitation to predict that they would is we just printed margins that were a couple --

Brian Lane

We've never done before.

William George

A couple of hundred basis points higher than we would have thought were possible a year ago. This stuff happens in the real world. And we literally -- we don't know how much it's going to rain, right? We don't know how other people around us -- you can't put wires in a wall that hasn't been built yet.
So we think that the environment is really good. We think that the people we work with are doing a great job of taking the right amount of work. Really, I don't know, we -- honestly, it's hard for us to say we can maintain a margin as high as we've had, but we looked at each other -- we all looked each other in the eye and said, really, I wouldn't bet against our guys. They're going to keep doing this. But it's not -- there's not a lot of science to it.

Brian Lane

We'll go for it.

Alex Dwyer

Okay. Got it. That's helpful. And then I just wanted to ask about how you think the trend in backlog could play out this year, like quarterly? And like should we expect a similar cadence to 2025 where like 1Q and 4Q are easier to achieve net bookings in the quarter? And then it could be tough again this year to sequentially grow the backlog through those heavy summer quarters?

William George

Yeah. So Brian and I were looking at a schedule this morning where our backlog jumped up 2 years ago in 2022. The burn was half 2 years ago in the fourth quarter, what it was in this quarter.
So part of what you have going on is we have attained such a high level and we're so busy. And there's a finite amount of sort of human capital, right, hours that can be worked and not kill people. And so our backlog, it's -- I would take the over on it. In general, over time, on average, but I just think the doubling kind of -- the crazy ups aren't there.
As far as the seasonality, if you ran Comfort, or a company like us for over 10 years, 8- of the 10-years, you're going to see more net backlog bookings in the first and fourth quarter. And 8- of the 10-years, you're going to see proportionately less more burn. It's just the way it is because of what time-space mass weather.
So I think if nothing's really changed, I think anybody who is watching our backlog and looking for signs of a slowdown should -- would be better off listing to us because as we said, there's so much work we're turning away that there really is no -- people look at our backlog and they worry about whether we'll have work.
And that -- if you're us looking at what we -- when we talk to somebody who does the backlog for a big electrical in Texas, that is so far from the world they're living in and the way they're thinking that you wouldn't believe it. It's like, yes.

Alex Dwyer

Okay. Got it. And then I guess my last question, I just wanted to kind of ask about your mix of business within the manufacturing portion of your revenue. And like as we think about that mix last year versus what's in the backlog and plan for 2025. Is there any -- is there any shift in those like sub-end markets between life sciences, chips and food beverage?
And then like if we get tariffs in addition to the reshoring theme and the [tips act] funding? Like how do you think that changes your slate of opportunities going forward at all? Like could it change the mix of new construction versus retrofit projects? Or does it not really change that much?

Brian Lane

Yeah. No, in terms of the -- in the first part of it, the mix is pretty consistent where we were last year with pharma, health care. We've got a few solar in there. But we have a lot of consumer product usage in there like you get a [legoland] job going. So there's a lot of mix of manufacturing that I think is going to increase these tariffs stick around that we'll be doing.

William George

Yeah. I mean it feels like on storing is -- you have to begin -- I mean it's really at the beginning. So there is a push. There's definitely a sort of a current towards industrial. But then people didn't quit putting (inaudible) on their sandwiches and people still want to feed their dogs, dog food and they still -- and even in the data center world, people sit here and they look at the new compute stuff, but people are still taking pictures and streaming, right?
It's not as if there are new incremental sources of demand, but I can't think of one that displaced an already existing source of demand. And that's why you're seeing some of the extraordinary outcomes that you're seeing is -- and then you put that on top of the fact that over the course of the financial crisis, and all the way back to [9-11] and then, of course, COVID, there was a lot of supply destruction, right? So you have unprecedented demand after 20 years of underinvestment in supply and we're all scrambling to catch up.

Operator

[Kadir Ritchie], [Ritchie] Capital Group.

Unidentified Participant

Thanks. Good morning. Thank you for taking my question. I hope you can stomach another pipeline question. But with the recent volatility around the AI space with the introduction of Deep Seek and other headlines. It's clear that the market seems to think that -- or view Comfort Systems almost as like an AI type stock. And I'm wondering if you feel like that characterization is fair. I understand the comment about impossible demand in the data center space.
But -- how do you make decisions about taking on more work in that sector versus other opportunities? Is it simply the highest margin per man hour? And so you take all you can? Or is there an approach to balancing on the backlog portfolio?

William George

Well, so two things I'd say. One -- so Comfort is [30] -- so we're 63% industrial, of that 33% is advanced technology. And of that, a little over two-third is data center. So Comfort is not a data center company, right? And -- low 20% of our revenue is data center.
And -- but having said that, the reality is how we picked jobs. Gross profit per hour worked, or really per $100 of labor, right, because not all labor is the same is the ultimate measure of how attractive a project is from the point of view of sort of profitability and spreadsheet considerations. But today, for most of our people, there's a prequalification process they go though where they say, will, this be a good job for my workers. Is there is there good parking. Does this GC run a good job and keep the field from being muddy.
I mean, because ultimately our workers, we've had them for decades, some of them are multigenerational. But they can get -- they can leave us and get a job instantly. And so our number 1, consideration, honestly, is, is this a good job for our people? Are the people they know going to be on it that they like to go to launch with? Is it with people we know treat people well.
The number 2, consideration is who are we doing the work for? The owner and especially the intermediaries like the GC and the other people on the job are they good people that we can trust, because we can afford to be picky right now.
And then the third consideration is, what's the gross profit sort of per hour worked or for the labor that we're limited by and that we're giving up for them. And the good news is those three considerations are not independent. Almost always the best answer to number1, and 2, will be the best answer to 3.
So the world we're living in is so different than what somebody sitting buying and selling stocks based on their guess is about AI, that it's very hard for us to really even answer that question.

Brian Lane

The other part that curve is we're mechanical electrical contractors. So our skills where their pipe finished sheet metal guys, plumbers electricians, we can work in any type of building that you want to build. So it's easily transferable. It's not some specialist skill that can only work in one industry will, at the end of the day, work in what the market dictates is available.

Unidentified Participant

Yes. I was actually surprised by it as well. But it really made me take a step back and think, okay, well, if there was something like a Deep Seek and it changed the fundamentals of the fundamental equation for AI and data centers. And so I was just -- it made me curious how insulated you are about the shift in spending patterns related to the hyperscalers and trees don't grow to the sky. And maybe this time it really is different.
But just -- is the demand that would replace that? Is it already there? Or would you have to go out and do some work if there was a big shift?

William George

Well before the words artificial intelligence has been mentioned on any conference call, on any -- for any public company in America, there was not an electrician in America that was (inaudible) around for something to do, right?
So the demand is there. Obviously, if the demand becomes frenzied the pricing gets better. There is -- so of course -- anybody who took economics 10, right, basic economics, supply and demand effect, the outcomes and if demand wanes, we don't really have a problem. We've been cash flow positive every single year we've ever existed, right. We've earned money every single year we've ever existed.
Some of those years were really bad years, right? They were after the year after [9-11], they were the worst year of COVID, the financial crisis, devastated nonresidential building, right? And at some point, is really not a question that we can be a good company and earn money. But obviously, we should be worth more if you think reshoring is real, onshoring is real, we are worth more. If you think people actually are going to need data centers, we're worth more.
The last thing I want to say about the data center stuff is, our whole experience has been -- and we've been building data centers since ['05, '06]. Our experience is when they find a way to do with less, or when they find a way to make more, they just want more. They don't say, okay, well, we have enough compute. Let's draw a line under that and move on to staring at our logo. You know what I mean, they don't -- the thing is they want more. And if they find a way to get more, guess what they want after that? More. It's a land grab, right?

Unidentified Participant

Yeah. That's great. My last comment is I'd really appreciate your loyalty to your employees. It's something that we've always admired about Comfort Systems and just how you fight for your employees. So congratulations on that, and congratulations on the quarter.

Operator

Thank you. I'm showing no further questions at this time. I would now like to turn it back to Brian Lane for closing remarks.

Brian Lane

All right. In closing, I really want to thank our amazing employees again. We're very grateful to you. Thank you, everyone, for joining the call today. We are very excited about the opportunities we're facing in 2025 and look forward to the year. Hope everyone has a great weekend, and thanks once again.

Operator

Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.

免責聲明:投資有風險,本文並非投資建議,以上內容不應被視為任何金融產品的購買或出售要約、建議或邀請,作者或其他用戶的任何相關討論、評論或帖子也不應被視為此類內容。本文僅供一般參考,不考慮您的個人投資目標、財務狀況或需求。TTM對信息的準確性和完整性不承擔任何責任或保證,投資者應自行研究並在投資前尋求專業建議。

熱議股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10