Quanta Services, Inc. ($PWR)

Okay. Hi, good afternoon, everyone. Thanks for joining us after lunch. So I am really excited to have a conversation with Quanta Services. And joining us from Quanta is Duke Austin, and also Jayshree Desai, who is the CFO.

And so we'll just like a fireside chat. I've got a lot of questions lined up. But for those of you in the room, I'm told there's a blue card here. You hit the QR code. And if you want your questions answered, I've got my trusty iPad, and I'll make sure that happens. But yes, let's -- first of all, thank you. Thank you for joining us. So the first question I have is on just like the portfolio. And so like over the last 5 years, Quanta has done a lot between like the M&A, there's been clearly like a big acceleration in the electrical infrastructure build out. And so the question I have is just talk about what you've become and why this version of Quanta is right for the one in hand.

Yes. Thanks, Chad. Thanks for having us. Thanks for the interest. Fundamentally, I think the core of Quanta is the same with the craft. And we've really worked hard to build a perpetual model within craft along all skill sets. But in saying that, I do think as we've evolved as a company and how we look today it's different. I mean I think we've transformed the company in many ways, and we've become something to the customer where it's a collaboration to get where they want to go. And we're really having discussions with -- at the client level, at the customer level, which is -- and even to make sure that it's not impacting their customer. So I do think it's different. I think we're in a different spot than we were in the past because of where we sit with craft. And I -- as we have this discussion really around capital budgets, timing, how they interface. I think people really when we have large load on one side and utility on the other, the dynamics are different. One is trying to go as fast as they possibly can and one is trying to make sure the ratepayer is in good shape. So it seems like a model you could -- you melt together and we're doing that. We're seeing those models where one willing to pay to make sure the rate payers, okay, and the other ones want to build. And we're in the middle of that nexus. And I think as we see it, we can provide tons of solutions around that. We can work with customers on their capital I think we're not hear our customers say we're going to use Quanta and that's how we're going to get our capital budget done. It means something to me. Years ago, you didn't hear that. And I think that's -- you hear that a lot now. We have common analysts as well and to hear on both sides. And I think that's really our goal is to make our customers more successful. If we do that, we become more successful so I think both sides from our large load technology type of TAM to our utility TAM and the renewables in the middle or generation in the middle of that. So yes, look, I think we're in a unique position, great markets I think we put the company in the right place from a strategy standpoint to grow with multiple ways to grow. So yes, I think we'll continue to compound earnings.

Okay. So how have all these changes translate your ability to increase the scope of work on a project? Maybe you can talk about maybe 5 years ago, which share of the project could you attack and then with some of the acquisitions that you've done and the investments that you made on the supply chain side and in areas like that, what does that look like now?

I think if we get to the utility TAM, in the past, generationally, it was primarily just kind of T&D build construction and we evolved that into some of the engineering, supply chain, now the generation of both renewable and fossil. So really, unless they're building nuke, which some of them are, most aren't that whole addressable market they're talking about is really a market that we can help with besides land or things like that. But that -- when you get into the 90% from 50 on that big addressable market, which I think we have a good moat there on the utility side that allows us to really lean into that become part of that industry and help so we've not only got ourselves in there, but we have allowed ourselves to take more of that position on the technology side. You saw a platform acquisition with Cupertino, it really got us into the customer base on technology. They were first movers there, had been with technology for a long time. And now we've built that out to where we've added the SI or mechanical platform as well as TriCity into that and then also our internal electrification that low-voltage electric that we put in there as well against that mode and really built that out. So we were just electric. Now we can do mechanical. We bought civil early last year. in January. So there's not much we can't do. Actually, we can do it all. It just depends on if we want to. So I do think we were finishing up balance of plant that is center today. We still perform 85% of it. So I think the same thing will exist. We're going to do it again. And more and more scope that continues. So as we see these big builds coming out. We'll be able to take that addressable market other than chips, we're not going to mess with those. It's not us. So we'll do the rest.

Got it. Okay. And just maybe sticking with M&A. So for as long as I've covered Quanta, you've had this great flywheel, right, acquired family businesses, reasonable multiples. But as the business has gotten larger, your deal size is going to need to increase to basically maintain the current level of growth and so the question I have is like how do you keep that flywheel going? What sort of infrastructure are you putting into place to ensure that you can kind of sustain like this pace of M&A to kind of maintain your growth rate?

Yes. Look, I don't think we have to transform the business. I think family businesses, the last 2 are at least 50 years old, I think one is 100, 20, 30 maybe. We see a lot of 100-year-old businesses. They want to perpetuate the name, they want to be a part of something different. So we don't really source deals, it's something we do it internally. We're not out trying to make acquisitions, they're coming to us, do they fit the model or not? What does the management team want to do? Are they staying or are they leaving? . So we want people to stay with us, stay with this long term and buy great companies. And I don't -- I think a fallacy is that we need to make acquisitions. We don't. It's inherent. I mean I look at it like it's my money, and it's your money. And I want to spend it in a way where I believe we can get the most growth and we can generate the most cash. So that's what we're trying to accomplish with acquisitions is to make sure we invest free cash in a proper way. And I think we've done that and the synergies are there when we make these acquisitions, we haven't bought companies today where, look, we see the data center market. But the companies that we acquired about 30% of the business was data center. We knew we could grow them because we knew the data center market on the other side, and we knew what they did. So it's really -- it's a built-in synergy within the business because we have it at the customer level. And we don't have to like press these deals to say there's a synergy. We're very disciplined about it. We owe it to our ownership to be disciplined. So I think we can continue to acquire great companies with great management teams that really fit our model. And I think we've done a nice job seeing markets, the vertical supply chain, the things that we've done there, we can see the markets and where they're going. As long as we can predict where things are going, we're able to lean into great businesses over time and yes, we've done a nice job.

Okay. So one through line that I am picking up on the M&A side is that you're leading more into like the mechanical, electrical, plumbing side of craft labor, can you just expand on that? And just like how you see that market evolving for Quanta, maybe from like an M&A perspective but also from just like a project-based perspective, right? We're hearing more about these large-scale mega projects and the growing MEP intensity. So maybe spend some time on that, please?

Yes. No. I mean it's a client. I mean, the client is asking us to do more. And so we're listening. And we need platform companies to do that. And so DSI would -- we felt like we needed to be in the mechanical business. It's craft. It's something that fairly we relate nicely, great family. We were able to really lean into that and put some capital in it and we're growing out, the fabrication facilities today. We're going to double the size of our fabrication and so that's been something that we're able to really meet the customer demand on that. And it's not just chips. I mean it's from Tesla to Samsung, I just think we've really been able to move that way past just the data center piece of it. And the craft, I think in the middle of that is just growing that craft. What they want is our ability to give curriculum training, the things that we've invested in since 2009. Management. That's my background. And so really being for generation is we've really put the time into craft. And now if we get someone into our facilities that can't climb because they're scared to heights. 100-foot is pretty tall. You've been up there. It will get a little shaky on you. A lot of other people want that. They get back down, about 30% fall out in the first poll. And so now instead of going home, they go and go to the inside electric or they go to pipe, the weld or plumber. We're able to really move people around and get a lot more people through our trade organizations -- and I see -- we see a lot of people that like that will say we'll give you $100 million. That's going to do with money doesn't buy craft. Craft builds craft, tournament builds tournament. You can't think you're going to put $100 million into anything today and get it craft tomorrow, it going to happen. It takes years to train and we're very proud of the adjournment that we have and the training we've done and it's a policy to think. We've been investing $200 million a year for the past, I don't know, since 2009 in training, and we're proud of that. So I would just say like not only are we making the acquisitions against it or training the people that allow us to get what I would consider the synergies in the outward growth of the company.

Got it. Okay. So let's actually shift over to talk more about year-on-year end markets. So both Duke and Jayshree, you spend a lot of time at the highest levels with clients, I'm talking about the utilities, developers, hyperscalers, co-locators. Can you bring us into those conversations? Like what are their concerns? Like what are they asking you for?

I'll let Jayshree go first, she sees more than me.

Yes. I mean I think it's most -- everyone is really excited about where they sit. It's -- I think it's a once-in-a-generation opportunity for a lot of our customers and for ourselves. So it's generally excitement is what we see. Now having said that, there are challenges that every part of our business is dealing with. You've got affordability concerns. You've got permitting challenges, you've got political challenges, and you've got the speed to market challenges. And at every level that we're working with, you got to be able to address all of those at the same time. And I think that's really a unique position that we can sit in because we have the ability to really affect in some areas more than others, but literally, we affect every one of those areas. And so our ability to help our customers weave through these dynamics has allowed us to really become more of a partner with every one of them and allow us to give you all the certainty of why we're confident about our 5 years and even beyond that and what we're seeing around this infrastructure build. It's really a question of what we're seeing is more when, not if, so how do we accelerate and get more certain around the win is what we're working with mostly with our customers. Well, I'll stop there and let Duke add more to that.

Look. We're in a different spot than we have been. And it's -- we have a sense of responsibility when people say we're going to use Quanta to build this capital and stick to our plan. I mean, I take it personally. So we're discussing it all the time. It showed up in the past year or so where I think the customer is asking us to change them as well. We're trying to change underneath us, where we're no longer just building out infrastructure. We're also being asked to look at it outward 5 years, make sure the supply chain is there, make sure we have right away, help with permitting, help with political, get us where we want to go and stay with me. Help me. And that discussion is a different discussion than answering RFQ. It's way different. And that's where we're at today is the discussion is more strategic. It's long. We're prudent about it when we look at it, I think it's really -- I commend for doing it. I do. I think it's something that to get out in front of something and recognize the fact that I'm not going to be able to get this done without help. I mean we have to collaborate ourselves. We're collaborating as well. We need their help to help them. So it's very much 360, the way we're thinking through it with clients today. And the more that we can get involved, the earlier we can get involved, the better we execute in that at the price and at the time frame.

Got it. And then if you look at your utility customers and then your large load customers, how far out are those conversations like extending to.

Yes. I mean, look, we're well past 2030 in places, the utility business, especially, I mean, it's out past 2030. The large load, I would say, even same -- those discussions are ongoing. I mean they're all in different phases. They're all in different time frames, but if you're looking at a combined cycle today, if you order it today, you're 30 months, 30, I don't know what they're saying today, but 30, 36 months, whatever they said, it's over 30 months. And then it takes a couple of years to build it, you're 5 years from today, if you order today. So any orders they have or if they book next quarter, they're 5 years from the next quarter. And so those bookings are pushing outward, so you can see it. You got to build transmission substations in. It's going to feed something so I just -- you have a lot of power today that's being done what I would consider a bridge. If you're building 100 megs at times, it's bridge power, it's going to go to the grid. At some point, you got to build a grid back to it. We're on both sides of that. We're building -- sometimes we're building intermittent power and have plans to build the grid. So look, I like what we said, I think it's much longer than 2030. I don't know why we put 2030 as the time frame, everything stops in 2030 world is going to end. Look, it's still moving forward like well past 2030.

Yes. So we track like the multiyear utility CapEx budgets, right? just only maybe like 1 or 2 years ago, you saw those budgets growing at like 8%, 9% CAGR. And now if you look at, at least like the last quarter, right, these multiyear budgets are growing at like plus 20%. So how do we think about that in the context of when that should translate into an acceleration in Quanta's utility business within electric power?

Okay. I think we're planning for that sure. Today, we see the growth rates in our people in the transmission business, we have a really nice business growing in a normalized way today. That's what it's doing. We have not seen the big project stack on it yet. I think you probably start booking into this year, early next year, start going to construction on some of them. But that's going to elongate. It's going to be broad. We've seen these cycles with 4 big ones, but I think it's bigger than normal. So the bigger projects will start to come in, in '27 and really stack on and you got 5-, 7-year build on just what you see. I think it's going to get bigger than go longer than that, of course. But I do think after that, we haven't -- we're in early innings on the bigger stuff that would stack on top. Same with our generation, so they're as well the same.

Okay. well, since you mentioned transmission. Let's jump there. So can you frame up the large transmission opportunity? And at least like from like the projects that we track, there's like a pretty big center of gravity in Texas, right? Like 1 thing I'm thinking about in Texas is that is probably a little bit easier from a permitting standpoint to get things done there. There may be a little bit more certainty. I don't -- new to me. But like -- yes, I guess like how do you think through the stacking of those sorts of projects?

Yes. I mean I would tell you Texas isn't till you goes through the wrong piece of property, it's great. You try to go through the wrong piece of property. There's no fun. But yes, look, you got -- we've got to build infrastructure. I think we're making progress regulatorily, Texas, it will start first, I believe, you're seeing MISO come out with projects that people are winning there. I think they continue to have tranches. There's some backlash on whether they weren't competitive or not a Delta go backwards, but they may. I do think it will be a blend of both. Both for the regulated utility to stay there, and then they'll have some competitive what they're doing now. It seems to be the way they're going to go, but we'll see. Either way, we're on both sides, it doesn't matter to us. We want to build it. So I do see those things moving forward in order to really facilitate kind of what we're trying to do as a country with the load. I mean we're going to double the load in the country. The need for transmission congestion is probably $12 billion to $15 billion a year, somewhere in there of congestion. It's the cheapest form of energy. So we've got to get these lines built. And I think you got to walk it back, and I know I hear a lot about, well, it's only 50% utilized. Well, the national highway systems 35% utilized. You're going to get rid of the highways too? Or you're not going to build any more, I mean, look, it never was -- the transfusions system was never meant to be 100%, ever. People go to sleep like something else sleep. I mean you got to sleep a little bit. And when you sleep, lights go out, everything goes everything is quiet, your bill goes way down -- of course, it's 50%. That it's not ever going to be more than that's a fallacy. So you've got out of that transition to get it to load centers. And we really need the permitting and the reform and some of the things to help that. You still have a lot at the state. The state has got to get behind it. I do think the vertically integrated utilities are probably in a better position in growth. Hopefully, we can do some regulatory things to give both sides of that, some ways to grow generation because we need the generation as well.

So is there anything that maybe makes you more optimistic on like the pace of the build from a regulatory standpoint, there's a recent acquisition of Dominion by NextEra. Like is there anything that kind of makes you a little more confident on that.

We have a lot of smart people trying to build generation transmission into PGM. We got -- I get some political pressure off and build it in. What I would say is demand is there. The willingness of technology to pay for incremental cost to ratepayers drive the ratepayers, philosophically under any scenario, more generation, more transmission should equal less to the rate payer. It just -- there's no reason why we can't do that. It's being done in Indiana. In Indiana, if you look at the model in Indiana, where NiSource is building there, it's $7 a month to the rate payer deduction because of loading. So it can be done. It's easier for the vertical people to do so. But definitely, I think you're going to see scenarios where you're seeing that model get pressed and so it will come down -- but we've got to have those kind of things happen in order for us to perpetuate where we want to go as a country. And that demand will allow us to do so.

So what makes 765 kV so much more complex than 300 or 500? And can you talk about just what makes Quanta just well positioned to win on that part of the cycle?

765 is big and heavy. And in some parts of the construction, you can go pick a block up at your hand. You got to have a 12-ton crane and you got to have the block to pick it up. And you got to pull wire across it. It's 6 -- conductor. It's just -- it's a different level of construction. I just -- when you think about it, it's big. And so a lot of capital that gets into that, a lot of planning that goes into it for us. You can -- 25% more right away for 6x load. I mean I sign up for that. Very hard to build D.C. across North America because it doesn't drop load. 765 allows you to drop load wherever you want so you can do it. You don't get as good as quality, but I won't debate that, sort of engineers. So -- but it is something that I believe like for a backbone infrastructure we needed in the country. It will allow us to really move generation much more economical across large territories. So I think it's going to get -- it's been talked about for -- I don't know in my career, 20 years. And it's always been needed. We just seeing now the demand is going to press it. And either we're going to build multiple corridors. -- which I don't think we'll do. So we're highly likely to see now 765 in a meaningful way across North America.

Okay. And can you talk about with this build-out of 765, like how that creates kind of like a follow-on like spur line opportunities. And so as you think about over the next 10 years, like this is kind of the first wave of building that trunk line. Like how does that cascade out as you're densifying that network?

Yes. I mean look, I think it looks just like a highway to go to highway because people understand you build 24-lane comes down to a loop or you have offshoots all the way once you get a corridor from point-to-point, and so all the lower voltage goes out to feed industrial parks or that whatever it would be, but you have to have the trunk line to do so. So it's 10:1 normally is what you see. So a big line you get 10% off of it normal. So it takes time, but over time, that's what you'll see.

Which just extends the capital bill when you think about it, right.

Text multiplier on.

If you had mile of 765, you got 10 of everything else, something like that give or take. So -- but I do think when you look at it, we see that. I mean, we see them planning to build those things. If you go back, Crest wasn't a 765, but it was 345 a circuit. The crest lines are at capacity. And we had the same debate should we build it. And we also, with the winter storm in Texas had a huge issue with CRE. So yes, we should build it. look, we have demand outside of data centers and we have demand with industrials. They have demand across the country with the pharmaceuticals. I can go on and on, industrial reindustrialization of North America. It's here and we got to wake up and start building the infrastructure.

Got it. That's actually good segue to the next question. I want to talk a little bit more about Cupertino and dynamic, 2 of your recent acquisitions. And just like how can Quanta, serve those -- serve a large load of customers better now versus before?

It's just a different discussion. I mean both of them are fabricating and different -- obviously, different disciplines. But the engineering, very common, the way we were able to kind of put them together, engineer together, do some things there. I think it's smart. We don't -- I don't think we're doing anything different other than scale, like we can scale across the lower 48 into Canada if we need to, but that scale of those businesses, we've added tied in there as well. So I think we're just able to see more and take on more of where they're at, build offices, build capabilities. We had a nice business that was more, I would call it, power plant industrial inside about 1,000 electric. We're taking those 1,000 and put them and do those -- frame into that framework on the electric side, the mechanical side, I think as you start to see generation get build from fossil, you'll see heavy mechanical that comes back meaningfully in these type of builds. So it gives us a lot of framework opportunity. we talk about all the time in the portfolio and how the labor is fungible. So you start to see, okay, neither 1 of these companies were really data center-driven they were doing all kinds of other things so you can see the fungibility of it. And now you have generation market coming on, it's going to start in a meaningful build on forward-looking. So I just think the opportunity for us to expand their capabilities are there -- and I like the fact that we're working -- they're both union, they're working together quite a bit. We're not having issues. I mean like type people, and they're both growing as fast as we can grow so really, really great management teams think alike, super proud of those acquisitions.

So maybe a question for you on the generation side. If memory serves me correctly, the vast majority of incremental generation like has been on the renewable side in the last number of years. But how do you see that on a go-forward basis in terms of the mix? And then I know with some of your I think with NiSource, right? You've talked about getting, I guess, back into gas-fired construction at scale selectively. So yes, maybe just on about like how you see that cycle playing out in terms of just that mix? And then how big does Quanta want to be in that cycle?

I mean, that's a good question. The last part. Got a choice, but -- what I would say, we are -- what I see in renewables, I mean Jayshree can comment, I think she should. You're seeing like we're still growing double-digit type growth in our renewable business. Batteries are coming into play. And what I'm seeing with batteries is we're seeing longer durations, and I'm seeing it get cheaper. I think Tesla kind of came out with something they believe is kind of revolutionary. I don't know. I haven't really looked up on it. But apparently, like you think -- can get that thing down in half, maybe to cost. I don't doubt him. If you can do that and get longer duration, solar becomes very much what I would consider as good as gas generation at times. So I mean, I see why not. So I just think, yes, you're going to have gas -- we need all phones, by the way, but when we see that, that's there, and that's what we're doing today. And as we progressed, we were asked to get into the gas-fired generation. It's not my favorite thing to do. But I knew we had the capabilities to do it. We just didn't -- it's not my favorite thing to do. It's very difficult to sink CGT up. We're getting unit 1 and 2 class some of the turbines, I know that's difficult. And there's very, very good people to do that. But they're busy. They're beyond busy. So the client is asking us to do more there. And I -- we had -- that work with us, had the capabilities, ran a big there's an engineering division before. We have 300 people in Birmingham today, going to 500, 600. And we're getting the front-end capabilities. We decided to do a joint venture on the first one in Indiana and we're able to do that, derisk the company through the construct, the way that we did it while really reducing cost to the client and the rate payer, the large customer to offset some risk. So I think that design is a good design. I think we can continue. And the inbounds on building gas-fired generation are probably every bit as robust as data center or anything else at this point. But we can't build them all, we're not going to try we're going to have clients that we really want to work with that are asking us to do something in a collaborative way, we'll do it. But as to be in a collaborative way that we can work together to derisk us as well as enhance the ability to come in on time and on budget. We have the craft. I thought early in my career that we built one in Alaska. I love them talking me into doing it. And man, it was tough. And sinking something that nameplate power is just hard. And so I know it's hard and that's why I'm very concerned and you hear me talk about it, how big can we get? We're only going to get as big as we feel comfortable. And that's one of the reasons that we devised ourselves with JV is because that's their main business at times -- they did a really nice job. And I felt like it's a great partnership on where we're at today. Single cycle is not too worried about. We can build single cycles. It's not something that is difficult as thinking them up. But look, lots of opportunity. And I do think it's really important when you think through because I get -- we get questions all the time on what if renewables go down. Well, what about our gas generation that we're building that business. And does it -- where does it sit? Because one thing I will say, we have great markets. They're not all going to rise at the same time and they're not going to stay risen forever. They're going to move up and down. And we'll have things that look great, things have looked good and things that are okay. And that's what we want in the portfolio. It allows us to really be flexible to give you long-term guidance, and derisk the investor base on 3 different verticals typically of initiatives the way the company is moving. So I -- look, if we hit them all at once and Jackpot, I guess, I don't know. Look, I don't think you'll hit them all at once, that going vertical on them. I just think there's going to be fluctuations, and it really gives us longevity in the markets that we see well past 2030. You should comment on renewables.

You do a great job of it, but not much more to add. I mean you know this, Chad, I'm clearly a bull on renewables, and I will always be because it makes sense. The cost curve is coming down. There is value to it in many markets and the power demand is there. And so we continue to see a really good market there. The technology curves on both those things. I mean I always say don't ever underestimate technology. And both solar panels and batteries are really technology plays and it just keeps getting better and better. So there's a real need for that as part of the energy mix. In the short term, for sure, because like Duke said, it's very difficult to get a gas plant built in the next 5 years. There will be some, but the majority of the generation buildout will still be renewals. And even beyond 2030, regardless what the politics might be around the tax credits, it will be about market demand and market drivers and where there's an energy need and even some capacity need, you're going to have batteries and solar be part of that. And wind is a little different. Wind is more challenge for sure, our onshore wind. It will still be part of the mix. It just won't be a big growth driver. But -- there's value to wind, actually on both sides of that equation on the energy and capacity side. And so there will be markets where especially as more transmission gets built out, which has been the biggest bottleneck to wind. You'll see more wind coming as well. But solar and batteries will still have a nice growth curve to it.

Got it. Okay. So how well behind the meter. I know right now, I think there's a view that it is maybe more like a bridge power solution, but maybe go beyond 2030, right? Think about what the grid looks like with that added element.

Yes. I mean I think there's cases that you can make where it makes some sense to have some form of meter generation. Look, I think batteries beyond the meter are going to be there for sure. There'll be other things back up generation, of course. So anything they're doing today will at least become backup generation. The question is, can you sell back into the grid, do you have bidirectional capabilities or not? What can you get out of that? Are we thinking through that well enough now of what can be done later. . So I do think there is obligations for it -- it becomes cost at some point, like if you can bring on load and you can bring on utility scale, it will reduce the rate payer. But it will -- like so you got to think through the cost of something. Right now, it's speed to market, that becomes cost to market over time as you start to get more generation on these systems, but there's always a place for it. We were investing in a company called Hibar they build rebars. We take some offtake off the rebar, green rebar in Arkansas. And they have behind-the-meter solar and battery to shape peak and what it does is for the price point of their electricity, they can build -- we can build and manufacture still rebar much cheaper because of the way that batteries in solar come into play and applying it. I'm amazed what that will do for a company that is press...

Back over to your modular fabrication capabilities. I mean, correct me if I'm wrong, but I think this is a recently acquired capability with Cupertino and dynamic. What have you learned about these sort of capability and then how do you think about deploying this like at greater scale? And if you can, can you talk about just what share of your work even flows through modular today?

I mean we had a -- if you take into account our vertical factories for transforms, you'll have 7 million square feet under roof with the -- we said we're going to spend $700 million on expansion. We're expansion, expanding all the factories. So that includes MEP fab capabilities. So as we look at it today, I mean, if you go back 10 years, Cupertino was a first mover in this. It's always been a good market. It's always been a market that we thought was prevalent. But in saying that, you didn't have the constraint of the workforce that you do today and the cost to move people, the cost to move people into other parts of the country. So it's much more economical today than it's ever been to modularize because of some constraints on crop scale. And especially the electric side. So I think those constraints -- you've always modularized some of the mechanical, but it's more prevalent now. So I think as you move into like rural areas, Wyoming, New Mexico, places like that where there's just nothing. I mean you're much more economic to fabricate. So we're seeing more and more of it. Your engineering, your video, your Trimble, all the things that you have capabilities of the 3D modularization that's really enhance what you can modularize and it actually fits when it hits the field. So I think that's the key is we've advanced technology to a point where you're able to have what I would consider world-class quality at your factory that goes to the field, everything kind of syncs up and it cuts that time to build down and it cuts labor down in the field. So yes, I like it. We'll use it as much as possible. I would tell you it's less than 5% of the business, call it.

Yes, roughly.

It's growing. That will -- once it expands, it will grow more. It's just -- the business is also getting bigger. So it's hard for me to characterize it against the business because it's just -- our business overall continues to grow.

Well, I'm going to have to ask about the economics.

Look, I mean, there's players in the industry that are pure play, you can look at them, and I mean our margins are comparable.

Okay. Okay. So you spent a number of years working with the utilities through like a master service agreement. It kind of backs the question with data centers with the long visibility. To what extent should we expect those sort of relationships start to unfold or even on like the industrial side?

Yes. I mean I think they are -- they're unfolding. We're definitely building on that side. And it's a trust issue. In many ways, like we got to continue to perform. If we continue to perform over time, it's already happening, where they realize when they sign -- when we sign up for something, we're going to do it on time and on budget. And the more times we do it, it just keeps building our name and our best commercial is our customer. I mean, I had one the other day. They were talking to one of our larger utility customers and they had given them our name and it wasn't a hyperscaler, it was a co-locator -- a large co-locator. And they said, "Hey, the utility came back and said, "Hey, they just thanked us for giving them your name. And I was like, oh yes. Thank you." So this -- it's very soon -- I mean like this thing is very circular. It's a small group, and we just got it perform, I think, and provide the same kind of mindset solutions that we have for utilities on the other side and just not deviate.

Yes. I think we work hard to be long term, and we work hard to be programmatic with our customers. I believe we've done a nice job of that on the utility side and you're seeing that in our results. We have been working towards that on the technology side. It's absolutely making an impact. It's just -- it's early. It's only been 2 years since Quanta really hasn't been leaning in on to the technology side as a Quanta hole, right? I mean, obviously, our operating units have done it for a longer period. But within this whole Quanta holistic total solution approach we've just begun. And we're seeing -- I believe we're going to see a lot of results of that as our technology customers are seeing the benefits of that. And like Duke said, we got to prove it because, again, they got to trust us to get what they need to do, and we're in the early innings in that.

Got you. Okay. So on the Investor Day, you guys talked about the $13 million per megawatt TAM for data center so I guess, like for like the average project that you're working on today, what does that look like? And what would you need to do organically or inorganically to actually realize that full $13 million TAM?

Yes. So like I think we look at it anywhere between, call it, 25% to 50% in that range the way we think through it. It's really -- depends on where it's at. Like regionally, just structurally, there are some regions of the country that were just not as heavy, and we don't have the capacity in that part of the world. So it's a matter of us building capacity in different areas, whether it be mechanical or civil or whatever. I mean the high-voltage piece we can cover off, medium voltage we can cover off, I can take it. But you got to think about it. We just started this like kind of this platform 2 years ago. And so we're building on this platform that's already big, big and so I think we'll just continue to add to it, add regionality to it, and then really pick the business up where it's more fulsome. I would tell you, we're probably averaging 15% now on things because we're building for others out of our facilities, fabrication facilities. We're just doing mechanical on one building versus 10, there's just stock kinds of things. I just -- we can take more market share, we can think through it. But I mean, we'll have top 10 customers there that we're going to service, and we're going to have to be more fulsome to do so. So we continue to talk to them years out, decades out around like what we need to do, and we're going to build against it.

Okay. So now shift over to labor. So can you talk about your strategy to ensure an adequate labor supply like over the next 3 to 5 years? And then specific to the training program, like what do you see is like the biggest bottleneck?

We've done a nice job since 2009, and I do think that continuation is there. I'm not as concerned about the craft, I mean I'm concerned, but I'm not as concerned. We have good, what I consider visibility and the growth curves and what we need from a labor curve. So we can see it today the more that we can get in front and plan with our clients on a forward look, the better off we'll service them. So I mean, we're encouraging, like, let's have the conversation early. Let's talk about what you want to do, where you want to do it. that moves all the time. And so like we can say we're going to go to Wisconsin, we end up an aisle. It happens, and that's okay. But as long as we have good growth curves with them, we can meet the demand. We're hiring a lot of military, about 25% of our craftsmen from military. I think we owe it to them. Our average -- our medium wage is #120,000. Kids coming out of school $75,000, 25% pension easily and, call it, all health and welfare for the family. I just think like we can do some unique things. We pay well. We always have -- I feel good about it. We push equity down to 9,500 people between 90,000 so we're proud of that as well. So we've got to take care of them. We got to take care of craft, it's who we are and our ability to ramp that. 5 years ago, we were kind of organically on the growth rate. Last year, it was 6,000 organically added 10, but that was similar to the acquisition. So -- journeyman we have more we can build, which should pathway, I feel good about where we can go at craft.

Okay. So maybe just the last minute or so. Let's talk through your margin targets, right? You call for an increase out to 2030. Can you just walk through the levers to get there? And since we're running short on time, I want to throw something in on return on invested capital. That's like another target for you. Maybe just talk through the building blocks there.

Yes. I mean, look, I think when you look at our margin profile, we're going it some. You can see the project mix inside the UI. It came up. I think that will continue as that becomes a bigger piece of the segment. But we've got a lot of -- we probably -- that's the question of the day margin. Look, we still perform 85%. We have to pay our craft. Our crop moves up, okay? We pay them well. It's going to come up. We are going to increase margins a little bit. But this -- that's not the story. We'll take margin -- we'll take outward look and longevity and stickiness with clients over time and compound earnings invest free cash and do great things with it that we've done in the past for a decade. When we give a 10-year, 5-year look, we hit it and we've hit it for a decade. And so that's the intent and that's how we do it. We just stay with it, stay and it's not very sexy, but you just compound it over and over again.

I mean it's the same approach, right? I mean our returns, you've seen it grow because of this consistent, deliberate effort on focusing on the long run. I mean there are levers. You've seen our working capital profile improve. We believe we're going to continue seeing that. That's going to be a driver of ROIC, the margin improvements that Duke talked about around some of the areas that we can do around the mix of work and the efficiencies we're going to continue to gain around those. All of those things are going to help drive ROIC. But we are heavily focused on ensuring that at the end of the day, we take on more share of wallet even if it might be not necessarily margin dilutive, but if we're allowed to take on more scope with less capital, we're going to do so, and that's going to be a big driver of our ROIC as well.

Got it. Okay. We're out of time guys. Thank you so much. Appreciate it.

Thanks.