STRL - NASDAQ

Good morning, ladies and gentlemen, and welcome to the Sterling Infrastructure Fourth Quarter and Full Year Webcast and Conference Call. [Operator Instructions]. As a reminder, this call is being recorded on Thursday, February 26, 2026. I would now like to turn the call over to Noelle Dilts, Vice President, Investor Relations and Corporate Strategy. Please go ahead.

Good morning to everyone joining us, and welcome to Sterling Infrastructure's Fourth Quarter and Full Year 2025 Earnings Conference Call and Webcast. I'm pleased to be here today to discuss our results with Joseph Cutillo, Sterling's Chief Executive Officer; and Nick Grindstaff, Sterling's Chief Financial Officer. Joe will open the call with an overview of the company and its performance in the quarter. Nick will then discuss our financial results and 2026 guidance, after which Joe will provide some additional commentary on our markets and outlook. We will then open the call up for questions. As a reminder, there are accompanying slides on the Investor Relations section of our website. These slides include details on our full year 2026 financial guidance. Before turning the call over to Joe, I'll read the safe harbor statement. The discussion today may include forward-looking statements. Actual results could differ materially from the statements made today. Please refer to Sterling's most recent 10-K and 10-Q filings for a more complete description of risk factors that could affect these projections and assumptions. The company assumes no obligations to update forward-looking statements as a result of new information, future events or otherwise. Please also note that management may reference EBITDA, adjusted EBITDA, adjusted net income or adjusted earnings per share on this call, which are all financial measures not recognized under U.S. GAAP. As required by SEC rules and regulations, these non-GAAP financial measures are reconciled to their most comparable GAAP financial measures in our earnings release issued yesterday afternoon. Our discussion of all results today, including revenue and backlog, refer to figures that adjust prior period results to conform to the current accounting of our RHB JV, unless otherwise noted. I'll now turn the call over to our CEO, Joe Cutillo.

Thanks, Noelle. Good morning, everyone, and thank you for joining Sterling's Fourth Quarter and Full Year 2025 Earnings Call. I'd like to start by thanking our team for delivering another outstanding year in 2025. We achieved strong revenue growth of over 32% and adjusted diluted EPS growth of over 53%. This is the fifth consecutive year we have achieved adjusted EPS growth of over 35%. Full year gross margins reached 23% and adjusted EBITDA margins exceeded 20% for the first time in our history. The strength of our margins reflect our continued focus on pursuing opportunities that offer the most attractive returns. Additionally, our operating cash generation remained strong at $440 million. We are pleased to discuss these results with you today, but even more excited about the opportunities ahead of us. The Sterling Way, which is our commitment to take care of our people, our environment, our investors and our communities, while we work to build America's infrastructure remains our guiding principle as we execute our strategy and grow the company. Moving to the fourth quarter results. Revenue grew 69%, fueled by 123% growth in E-Infrastructure Solutions and 24% growth in our Transportation Solutions. Organic growth in the quarter was 36%. We grew adjusted earnings per share by 78% to $3.08 and adjusted EBITDA by 70% to $142 million. Additionally, operating cash flow generation in the quarter was again very strong at $186 million. Our backlog position and strong visibility drive our confidence in the future. Signed backlog at the end of the quarter totaled $3 billion, a 78% increase from year-end 2024. On a same-store basis, backlog increased approximately 50%. When you layer in our unsigned awards of $301 million, and pipeline of future phase opportunities, which now exceeds $1 billion, we have visibility into a pool of work approaching $4.5 billion for Sterling. Now I'd like to discuss our segment results for the full year and fourth quarter in more detail. In E-Infrastructure, full year revenue grew 59% including 40% organic growth and adjusted operating income grew 67%. Adjusted operating margins reached nearly 25% and an increase of more than 120 basis points. This was driven by our shift towards large mission-critical projects where our superior project management and ability to finish jobs on or ahead of schedule is extremely valuable to our customers. In the fourth quarter, revenue grew 123%, including 67% organic growth. The data center market, again, was the primary growth driver in the quarter. Additionally, our geographic expansion efforts are really paying off. Our Rocky Mountain site development operation, which is solely focused on mission-critical work, grew more than 150% from the prior year period. Adjusted E-Infrastructure operating income grew 91%. Operating margin for the legacy E-Infrastructure site development business were flat with prior year levels. Margins continue to benefit from our focus on large mission-critical projects and strong execution. The CEC acquisition is performing very well. In the quarter, CEC revenue increased 21% from its prior year fourth quarter, and margins were in line with our expectations. The Texas market, in particular, is very strong, and we continue to see tremendous opportunities ahead for both electrical and site development. Our multiyear visibility and the E-Infrastructure business remains excellent. The aggregate of our E-Infrastructure signed backlog, unsigned electrical awards and future phased site development opportunities totaled more than $3 billion. Mission-critical work, including data centers, large manufacturing projects and semiconductor represented 84% of E-Infrastructure signed backlog at the end of the year. Future phase work is predominantly related to mission-critical projects. Moving to Transportation Solutions. For the full year, revenue grew 17% and adjusted operating profit grew 66%, driven by strong market demand and the benefit of mix towards higher margin services. Fourth quarter revenue grew 24% and adjusted operating profit grew over 100%. We ended the quarter with Transportation Solutions backlog at $1.1 billion, and 81% year-over-year increase, driven by strong award activity and the conversion of unsigned backlog to sign backlog. Shifting to Building Solutions, full year revenue declined 6%, and adjusted operating profit declined 23%. In the fourth quarter, segment revenue declined 9% and adjusted operating margins were 10%. Overall demand for homes has been impacted as potential buyers struggle with affordability challenges. Even with the headwinds in Building Solutions, the strength of Sterling's diversified portfolio and strategy to focus on growth in high-margin end markets enabled us to deliver another fantastic year. With that, I'd like to turn it over to Nick to give you more details on some of our financial metrics and the 2026 guidance. Nick?

Thanks, Joe, and good morning. I'll begin with our consolidated backlog metrics. Our year-end backlog totaled $3 billion, a 78% increase from year-end 2024 or 49%, excluding CEC. Combined backlog of $3.3 billion increased 81% from prior year-end or 42%, excluding CEC. Fourth quarter 2025 book-to-burn ratios were 1.64x for backlog and 0.81x for combined backlog. Moving to our cash flow metrics. Cash flow from operating activities for 2025 was a strong $440 million. We expect continued strength in operating cash flow in 2026 in both the legacy and recently acquired businesses. Cash flow used in investing activities for 2025 included $77 million of CapEx and $482 million for acquisitions, including CEC. For 2026, we are forecasting CapEx in the range of $100 million to $110 million, with the increase driven by investments to support growth and productivity. Cash flow from financing activities was $162 million outflow, primarily driven by share repurchases of $74 million at an average price of $168.72 per share. In the quarter, we deployed $26 million into share repurchases at an average price of $310.09 per share. Remaining availability under the existing repurchase authorization is $374 million. We will remain opportunistic in our approach to share repurchases. We are in great shape from a balance sheet perspective. We ended the quarter with $391 million of cash and debt of $291 million for a cash net of debt balance of $100 million. Additionally, our $150 million revolving credit facility remained undrawn during the period. Given our strong liquidity, we are in an excellent position to continue to take advantage of both organic and inorganic growth opportunities in the years ahead. Now I'd like to discuss our guidance. Our current backlog, visibility and strong market tailwinds position us well for another great year ahead. We are initiating the following guidance ranges for 2026. Revenue of $3.05 billion to $3.2 billion; diluted EPS of $11.65 to $12.25. Adjusted diluted EPS of $13.45 to $14.05, EBITDA of $587 million to $620 million. Adjusted EBITDA of $626 million to $659 million. The midpoints of these ranges reflect strong year-over-year growth of 25% or higher for each of these metrics. Now I will turn the call back to Joe.

Thanks, Nick. As we look to the future, we remain very bullish on the multiyear opportunity in each of our markets. Our strong backlog, future phase opportunities and conversations with our core customers on the size and number of future projects contribute to our confidence. In E-Infrastructure, site development, we anticipate that the current strength in data center demand will continue for the foreseeable future. We have discussed for some time that we're in conversations with our customers regarding how we can best support their strong multiyear capital deployment programs. As part of this, we are getting pulled more rapidly into new geographies, including Texas and the Pacific Northwest. Additionally, our projects are getting larger and are spanning longer time periods. In the semiconductor and manufacturing markets, there remains a very big pool of mega projects on the horizon for the later part of the decade. This would include planned semiconductor fabrication facilities. We believe that some of these projects are close to shore and will be awarded in 2026. In E-Commerce distribution, award strengthened significantly in 2025 as large customers restarted their investment programs following the post-COVID build-out and correction. We believe that, that momentum will continue into 2026 as these programs accelerate. Together, these dynamics across our end markets support strong growth opportunities over a multiyear period. We have very good momentum in our Electrical business for 2026. Customer demand is very strong, particularly in the Texas data center market. We have a high degree of confidence in our ability to leverage the combination of our site development and electrical services to drive growth and margin expansion across the platform. For 2026, we expect to deliver E-Infrastructure revenue growth of 40% or higher. This includes 20% growth or higher in the legacy business. Adjusted operating profit margins for the E-infrastructure are expected to be in the 23% to 24% range. In Transportation Solutions, we're in the final year of the current federal funding cycle, which concludes in September of 2026. We have built over 2 years of backlog and continue to see good levels of bid activity. For 2026, we anticipate continued growth in our core Rocky Mountain market. The downsizing of our low bid heavy highway business in Texas is progressing according to plan, resulting in some moderation of Transportation Solutions top line and backlog which should continue to drive margin improvement as we move through the year. We expect Transportation Solutions revenue growth in the low to mid-single digits in 2026 and continued margin expansion. In Building Solutions, we believe the business is well positioned for growth over a multiyear period. Our key geographies of Dallas Fort Worth, Houston and Phoenix are all expected to see population growth, driving new home demand. Additionally, there is an opportunity for share gain coming out of a down cycle. In the near term, we believe the current soft market conditions will continue. We anticipate that Building Solutions revenue will decline in the high single to low double digits in 2026 and that adjusted operating margins will remain in the low double digits. As a reminder, from a seasonality perspective, our fourth quarter and first quarter tend to be slower than our second and third quarter, with the first quarter typically our lowest of the year. On the acquisition front, we are continuing to look for acquisitions that are the right strategic fit to enhance our service offering and geographic footprint. We are seeing more high-quality acquisition targets in the market today than we did a year ago. Moving to our full year 2026 guidance. The midpoint of our guidance range would represent 25% revenue growth, 26% adjusted EPS growth and 28% adjusted EBITDA growth. With that, I'd like to turn it over for questions.

[Operator Instructions] And your first question comes from Brian Brophy from Stifel.

I guess I just wanted to ask about transportation awards and backlog. It seems like it was a lot stronger than folks were expecting. Anything notable to call out there? Were there any large awards worth highlighting?

Yes. Thanks, Brian. Nothing in particular. There wasn't one big giant project or anything. I think what people tend to get confused with is even though we're coming to the end of a funding cycle, only about 50% to 60% of the total funding has been spent. So we will continue to see good bid activity through that September time frame as projects are continuing to be let, which will obviously be built for a time period after that. So we feel very good of where we're positioned, and we like the bid activity. Nothing major, just kind of good singles, doubles and maybe a little triple now and then as we go forward. But I also want to just remind you and everybody that at the end of the funding cycle, the world doesn't stop. This isn't a toggle switch. Generally, what happens if they don't have the next funding bill in place is they will do an extension of the existing funding bill. They have generally historically adjusted that for inflation. And as a result, bids will continue to come up.

Yes, that's helpful. And then I guess just as a follow-up, you touched on this a bit in your opening comments, but hoping you could expand a little bit more. But just an update on some of the progress entering in Texas on the site prep side? And how is progress going kind of marrying that with some of the joint awards at CEC?

Yes. We're really excited about the Texas market, one from the CEC side on what we're seeing for data center expansion, electrical. But two, we're getting pulled very rapidly on the site development side. And I think in the first half of this year, we're going to be able to talk about some very nice awards that take place in Texas. And we have been, I will tell you what I believe, ahead of schedule and more optimistic about the strength of putting these 2 together and the responses that we're receiving from customers. So we're very pleased on the thesis is coming true. And I think we'll see some very good activity in the first half of this year that will make everybody happy.

And your next question comes from Brent Thielman from Davidson.

Great quarter, guys. Joe, maybe just to tack on to Brian's question, to get out a little bit more about how the pipeline has evolved at CEC since you've acquired it? Are the award opportunities getting a lot bigger? And maybe if you could comment on kind of what you're doing or seeing already that might support a higher threshold for margins for that business going forward?

Yes. We are -- the jobs are getting a lot bigger. These data centers in Texas, we were joking at our recent leadership meeting our team said, remember when we were so excited, we got the first 100-acre data center. And this is just a few years ago. And we said, yes, we thought it was the biggest project and all that kind of stuff. And they said, yes, we just started one in Texas and the parking lot is 100 acres, right, kind of put it in perspective. These things keep getting larger and larger, are having more and more -- these aren't data centers anymore, they're data campuses. I think people get confused about that, and we'll have multiple buildings on them. CEC has got great traction and combined, the site development CEC are getting great traction. We see margin improvements in a couple of areas. Obviously, as CEC moves more of their mix shift towards data center, that margin is historically better than the smaller kind of industrial commercial jobs. But also as we're combining the exterior electric with the site development. We have already seen significant margin improvements with the small dry utility business we bought in Georgia last year. And the exact same thing will happen with CEC over a period of time. So we're in the early phases of getting into those projects. That's where I say, I think we'll see some great progress in the first half of this year, and we'll start seeing some of that impact as we get to the second half of this year.

Okay. All right. And then maybe just on the legacy site development margins, as you mentioned, sort of flat here, obviously, at great levels. Maybe, Joe, you could just comment on why we shouldn't think they peaked at this point?

Sorry, Brent, I misheard you. You're asking about site margin -- site development margins in the quarter. We don't see them going negative. That's for sure, especially as we're seeing larger and larger jobs come out. We're also seeing some opportunities in the Northeast for some larger jobs that will help their margins improve significantly. One of the things we'll begin talking about more and more -- as we said in the script, our Rocky Mountain site development business grew 150% year-over-year. We're really excited about that. But one of the things we have that hurts our margins out in that area is we tend to have a little bit smaller and different equipment suite than we use on the East Coast. So we're going to be investing in some capital and doing some different capital planning that we think we can drive their margins up a fair amount as well, just by duplicating exactly what we do in the Southeast and the East Coast. And then strategically, we'll continue to look at more vertical integration. There's some leverage and productivity we can get out of vertical integration. As we move to a new geography, that's not where we start. We kind of get our feet wet, get into the site development, execute that gives us time to evaluate some of those players around the vertical integration, pick the best ones. And you'll see over the next 12 to 18 months, us tucking in some of those around the U.S. I do want to add one more comment, though, on your CEC question on how rapid they're growing and everything else. I would tell you if I had another 1,000 electricians in Texas, and I have a much bigger number. I think we could put them to work in 30 days or less. But when we bought CEC -- they had a small modular build facility. We just are in the progress -- the process of signing a lease that triple the size of that. That will also help their margins so we can start prebuilding more in a factory condition and ship those to the field.

And your next question comes from Manish Somaiya from Cantor.

Congrats on the quarter. A couple of questions for me. Maybe beginning with Joe, you talked about $1 billion of high probability future phase work. Can you give us a sense if that's tied to existing customers or programs? Or is it completely new? And how should we think about the expected conversion window?

Well, first, that $1 billion-plus is tied to projects we're actively working on today. So that's real work. That's what it's going to take in a minimum to finish out those projects. If you remember, as they continue to design, they release that work in a package. We did that next phase of work and then we execute it. So it's very solid. We can't technically call it backlog, but internally for capacity planning and everything else, it's a backlog for us. It's tied to existing customers, but we're always getting a new customer here and there. So we're not just fixated on 1 or 2. But the lion's share of that are with the big name hyperscalers that everybody knows that have all announced their budgets, plans and future forecasts, which are all up significantly. I will tell you, as we look ahead in the next 3 to 5 years, and work with them on their planning. We don't see anything slowing down. If anything, we continue to see it accelerate.

And then a question for Nick. As you guys scale up the [ E-Infrastructure ] business, how should we think about investments in working capital and free cash flow conversion?

Yes. So we expect to continue to have strong free cash flow conversion. I mean, conservatively, we're seeing free cash flow version in the -- to EBITDA in the 80% range or so and that's a conservative number. And we expect that to continue as we move further into the E-Infrastructure space throughout the year.

Okay. And then back to Joe. As we think about capital allocation, Joe, you mentioned obviously, you have an active share buyback program, $400 million initiated back in November of last year. You talked about M&A. You're talking about good assets coming on the market, which is interesting because I haven't heard that in a while. Obviously, multiples are high. But maybe if you can just give us your sense of your priorities. Is that a fourth leg that we should be thinking about to the Sterling business? So how do you sort of think about allocation -- capital allocation, M&A, fourth leg and why this whole abundance of good acquisitions because we've heard that on a couple of calls, and it's been a little bit surprising. But if you can please help us understand that.

Yes, our focus won't be necessarily on a fourth leg. It's really around when we step back and we look at data centers, semiconductors, pharmas coming some other manufacturing that we see along with longer-term projects. Right now, for the next 5 to 7 years, the best returns for us are going to be how do we grow, expand good services and geographic footprint within the infrastructure. Now that's not to say we won't -- if the right fourth leg came along, we wouldn't aggressively go after it. But right now, the opportunities that are ahead of us and what our core customers are asking us for in asking locations that go to services to provide for the next 3 to 5 years, really has us focused on continuing to look for that geographic expansion of site development and then also incremental electrical footprint and services and skills, and we will be looking at mechanical along the way as well as kind of a natural progression. So that's where we're focusing. Why do I think there's more quality businesses on the market. I think 2 things, multiples are certainly up some on the businesses. But two, I think a lot of these owners see this tremendous opportunity ahead of them, and they're not able to capitally fund the growth and they feel like they're going to miss the opportunity. And frankly, if they don't team up with somebody big, they're going to be on the outside looking into this and they're going to ultimately lose that work. So they've been around this world a long time, and this is something that none of us have seen in our career from a standpoint of the -- it's not billions. I think it will get the trillions of dollars of infrastructure spend that's coming in front of us that whether it's our yellow iron or our electrical skills or some of our other things that write into.

And your next question comes from Alex Rygiel from Texas Capital.

When you talk about manufacturing and high-tech and fab plants, can you just sort of help us to maybe understand how big that market opportunity could be sort of in the coming years?

Well, when you -- let's start -- we can start with the semiconductor stuff. And I'm a believer that we are -- we're at the early innings of semiconductor. I don't think we're going to see a major phase of semiconductor come on until 2030 or so just with the length of time of the announced capital spend that it takes to get a project ready to break ground. But in the meantime, there are a few projects that are going to be hitting here in the near future. Put it in perspective, a data center job for us is kind of a 3-year average, I would say. A semiconductor plant that are coming out will be closer to 7- to 10-year projects instead of hundreds of millions of dollars, the total scope of those projects could approach $1 billion. So that's pretty sizable for us. The pharma plants have announced multiple facilities around the country. Again, don't get confused when they announce it from the time they announce to the time anything happens, it's going to be 3 to 5 years. It's just the cycle by the time you get the land, you permit it, you get utilities run. They have to purchase equipment. Usually, that's specialty equipment and it's not a quick turnaround. So there's anywhere from 2 to 5 years of lead times depending on the parts and components of the industry. So we see all of that coming. And in parallel, I would tell you, our data center customers over that same time horizon are not slowing down. They're talking about bigger builds, more builds and working with us to try to commit capacity and capabilities. But also when you see our geographic expansion taking place over the next 12 to 18 months, it's not necessarily because there's something there today. It's because we know something is coming in the future, and we're working with our customers to be prepared for that. And the Texas market is on fire to say the least. It is unbelievable what is happening in Texas. And we're attacking that from the East, and we're attacking from the West. And then we've got our CEC business right in the middle of Dallas. We feel very, very good about what Texas will add to our company over the next 3 to 5 years.

And your next question comes from Louie DiPalma from William Blair.

Joe, you previously on past calls, you've discussed Sterling's entry into West Texas. Are you able to further expand into Texas to cover some other key markets? I know you said that you're attacking it from the East and attacking it from the West. Have you already won jobs on the East side of Texas and in other markets across the state?

Yes. When we -- let me talk about the attacking from the East, attacking from the West and then the market, okay, just to make sure I don't create more confusion. So West Texas is growing very rapidly. And I would tell you, we're actively working and winning jobs in West Texas that we'll be able to announce in the first quarter. What we consider East is kind of Dallas, Houston, that corridor, not East Texas is the piny forest. There's not a lot going on there yet. Maybe there will be. But when I say we're attacking it from the East and attacking it from the West, we're using our plateau business to come and start hitting what we call East Texas. So that Dallas market, there's stuff going on up in Oklahoma in the near future that's coming out. And then believe it or not, what most people don't realize is it's almost the same distance from Houston to West Texas as it is from Salt Lake City to West Texas. That's how big the state is. We're attacking the West using our Rocky Mountain resources and assets. And we'll be looking for strategic acquisitions obviously, within Texas or closer proximity to continue to add assets and resources and capability and capacity. Does that help?

Yes. That was great, Joe. And so should investors view 2026 as a year for -- in which you're focused in terms of your geographic expansion, you're focused mostly on Texas? Or are there other data center and mission-critical geographies outside of the Southeast and outside of Texas that you can potentially expand into for 2026?

Yes. I think in 2026, the lion's share of the data center growth will come in those areas. But there's some good activity taking place now in the Northeast. It's early. We're excited about some projects up there. We will -- the 2 other markets that are out there in one, I think, in 2026, you'll see us start moving more assets and resources to the Pacific Northwest. And again, there are several projects that will probably be released in '27 up there that could be very exciting for us. . The market we haven't figured out how to get into yet is kind of the Ohio, Indiana market. That's always a potential for us. I won't tell you we're on the 10-yard line of getting in there or anything else but that's another fairly sizable market. But when I step back and I just look at the Southeast, a few of the projects up in the Northeast and Texas, and you got the lion's share of the new stuff coming out, at least with our core customers and the size of those projects really fit our profile versus when you get into mid-Atlantic there's a lot of number of data centers, but they're small and just don't necessarily fit our profile.

And one final one related to this discussion for your entry into new markets. Is it reasonable to assume that the profitability in terms of the operating margins will be lower than like the profitability for your core E-Infrastructure mission-critical business such that as investors and the sell-side as we're modeling should there be any negative impact from expanding into Texas and expanding into the Northwest? Or how should we consider that?

Yes. I think depending on if we move equipment suite as an example from the Southeast to Texas, we're going to see the exact same margins. We've seen a little lower margins, the same pricing, same kind of process, a little bit lower margins in our Rocky Mountain, not because of the market just because we're ramping up that equipment profile. Remember, we kind of converted highway assets plus some large assets to get them underway as we continue to build that out, those margins will grow. Where we would see -- where I think we will see when we look at acquisitions, as an example, most acquisitions, margins are a little bit lower than ours, right? But we believe that after we get those acquisitions in our footprint, introduce our processes, our procedures, our technology and do those sort of things, we can continue to grow those margins up to where we are. And yes, just the other thing to keep in mind, Louie is our margins -- I'm sorry, our projects remember, the multiphase is anywhere 3, 5 7. They always start out at a lower margin and those margins improved based on productivity and what we're able to do staging earlier phases to help productivity on the future basis.

Your next question comes from Adam Thalhimer from Thomson Davidson (sic) [Davis].

Nice quarter. Joe, can you talk more about the CEC modular expansion? Curious what that will take you to on a square footage basis? And what exactly are you doing modularly versus in the field?

Yes. So our simple take on that is we're not the only ones doing modular, but anything we can take out of the field in prefab to move into the field reduces the number of electricians we need on that site, right? But it also improves productivity and cost. So the new facility is over 300,000 square feet. I will tell you that we're looking at does it or does it make sense to have multiple facilities throughout the U.S. that can make these components. So we're doing everything from the exterior piping and conduit around the duct banks to the actual cabinets that go into these centers. And we're looking at how do we continue to expand that capability and growth. It's pretty exciting because it really -- when you run the numbers, it's significant on what it frees up for capacity and how it positively impacts your margins.

Future of construction.

What's that? .

I guess the future of construction.

Yes, I think we will see, regardless of us or our space or other spaces, you're going to continue to see modular activities grow, is the labor pool just gets tighter and tighter. It's the most logical thing to do.

And then just lastly, I wanted to see where your head was at on the residential bid. And does it make sense to do an acquisition there when multiples might be more depressed? Or does it just make sense to let that play out?

Yes. And we don't -- I'd like to tell you there is light on the horizon. I think it's going to be tough, certainly first half of 2026. Things are just -- there's nothing positive that's happening to spark or drive the residential business to turn around and I'll also warn everybody that once it does turn around, there's 3 or 4 months of inventory on the ground that these builders are going to sell before we start seeing new builds start, right? I think that's just a fact. But you're not far off. I will tell you, if the right acquisition came along in residential and multiples are down. And by the way, their earnings are down we feel optimistic long term because the last time when we went back and looked at Tealstone coming out of the last downturn, they picked up significant market share on the back half of this, and we think we'll pick up market share coming out of this, especially in the Houston and Arizona markets. So if we found the right acquisitions, we would shy away from them, but we get them at a huge discount. And we look really smart 18 months from now.

And your last question comes from Julio Romero from Sidoti & Company.

I wanted to stay on the topic of these projects getting bigger. As you said, Joe, they're not even data centers anymore. They're data center campuses. As these mission-critical projects get bigger, more complex, is the mix of above-ground work versus underground infrastructure changing at all? And if so, is that mix shift margin accretive?

Yes. We -- I wouldn't say we've seen any significant shift on above ground versus below ground. I think what we will see kind of the next generation of these projects are going to have self-power generation on them. That will create more development opportunities on these because if you have a power plant, you got all the underground, whether it's water utilities, you got the electrical coming in, all that stuff. So we think that the size and scope of these projects, not only in physical size will get bigger, but the amount of stuff that we will touch will continue to grow. Those projects really won't start getting until '27 or '28. We may see one this year, but that's the lion's share. But it also opens up opportunities to look at other goods and services that we can add to those projects.

Got it. And does that change at all if the mix within mission-critical begins -- does skew over time to semiconductor and advanced manufacturing?

Yes. No, there's not -- there's not a tremendous difference. Obviously, you got duct banks in the data centers, but there's all -- what people don't realize is any of these facilities, literally has thousands of miles of underground pipe and wire and everything else. It's an underground city. So for us, we're not concerned if the mix shifts more towards semiconductors and away from data centers or more towards manufacturing. We've historically seen -- we've got more data on the site development side, obviously, but we've seen comparable margins. That's the beauty of the model we have. And now with electrical and as we continue to expand the footprint of electrical and capabilities, our stuff is fungible. We really -- we -- big projects, obviously, are much better for us, but we don't really care if it's a data center or a chip plant or a large automotive facility or a battery plan.

Very helpful. And I wanted to ask how Sterling is positioned amongst other specialty contractors with regards to AI-driven tools as these tools kind of become more adopted across the industry. Joe, how are you thinking about staying ahead of competitors? I'm sure some of those competitors would say they're probably using AI to narrow the gap of reliability or maybe scale versus Sterling. How do you see Sterling position in that context? And could you be using these tools to maybe widen your differentiation versus some of these peers?

Yes. Well, I certainly would not be able to see. I don't know where our peers are. I can tell you where what we are. I think people would be shocked what we've done and what we're doing in and around AI and how we're tying that to not only the estimating and bid process, but the project execution. We did 3 pilots last year. To say the teams were excited when we first -- when we first started them, everybody said, how is AI applicable to what we do. Remember, we're running drones. We're running all kinds of analytics of our equipment. We're now tying all that into project management and making things happen faster. We picked up just in, remember, our capacity constraint in site development is project managers. And we picked up somewhere between 15% to 20% of incremental capacity on project managers just from our first AI project. We've got 6 AI projects underway right now. So I will tell you, it's when people talk about, well, is AI real and all that, I would say if guys like us are running very quickly at this and our guys, every time they touch it, they come up with 4 or 5 other things that we can incorporate which will improve the efficiency, the effectiveness, the quality, and just as importantly, there's some really cool things we're doing on the safety side with this that will help make our employees even safer than they are today.

And there are no further questions at this time. Mr. Joe Cutillo, you may continue.

Thank you. I want to thank everybody for joining today's call. If you have any follow-up questions, please contact Noelle Dilts. Her contact information is in the press release. I appreciate it, I appreciate what our team did. Another great quarter, and have a great day.