As the utility-scale solar market collides with an era defined by massive load growth, EPC (engineering, procurement, and construction) firms are rethinking their strategy to meet the moment.
In this episode, Shayle speaks to George Hershman, CEO of SOLV Energy, one of the largest solar and storage construction firms in the US. George offers a unique perspective into the state of the market as well as the logistics of building gigawatt-scale projects and insights into how automation is changing the EPC game.
Shayle and George discuss:
- Why George believes rising demand can help solar move past boom-and-bust cycles
- How SOLV is taking on larger projects without needing to increase its workforce proportionally
- How automation helps SOLV build and install utility-scale solar faster
- The logistics bottleneck impacting EPCs’ ability to scale
- How AI-driven simulations can help optimize installations
Resources
- Catalyst: Can AI revolutionize EPC?
- Catalyst: 2026 trends: Gas turbines, Texas’ load queue, and China electrifies
- Catalyst: Scaling America’s domestic solar supply chain
- Latitude Media: Can the US bring solar installation to below $2 per watt?
- Latitude Media: This former solar installer is all-in on software-only sales
Credits: Hosted by Shayle Kann. Produced and edited by Max Savage Levenson. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.
Catalyst is brought to you by FischTank PR, an award-winning climate and energy tech, renewables, and sustainability-focused PR firm dedicated to elevating the work of both early-stage and established companies. Learn more about their PR approach and how they can support your company’s messaging by visiting fischtankpr.com.
Catalyst is brought to you by EnergyHub. EnergyHub helps utilities build next-generation virtual power plants that unlock reliable flexibility at every level of the grid. See how EnergyHub helps unlock the power of flexibility at scale, and deliver more value through cross-DER dispatch with their leading Edge DERMS platform, by visiting energyhub.com.
Tune into Critical Capital, a brand new podcast from Crux and Latitude Studios. Hosted by Crux CEO Alfred Johnson, Critical Capital explores the interlocking forces powering clean and critical infrastructure. Join us every other Tuesday for in-depth conversations at the intersection of energy, government, finance, and global markets. Listen here, or wherever you get podcasts.
Transcript
Shayle Kann: I’m Shayle Kann. I lead the early stage venture strategy at Energy Impact Partners. Welcome to Catalyst. Well, I’ve spent enough time on this podcast talking about EPCs or engineering procurement and construction firms for those not in the know. Anyway, it’s time to talk to an EPC. SOLV Energy is one of the biggest, particularly in the construction of solar and storage projects, depending on your metrics, Solve is usually either the biggest or the second biggest in the US. Anyway, they’re big and they build a lot of big projects. They also went public earlier this year in a very successful IPO and somehow managed to snag the ticker MWH or megawatt hour, which I very much appreciate. Anyway, George Hershman, SOLV’s CEO, has a really fantastic window into a bunch of things I’m interested in, ranging from the state of the market to the labor shortage, to the extent that it exists, to the impact of AI and robotics on construction.
So here’s George. George, welcome.
George Hershman: Well, thanks for having us or me, Shayle. Appreciate it.
Shayle Kann: Excited to chat with you about the state of EPC in power and solar and storage in particular. And I’ll maybe ask you to start by giving me your state of the market. Where are we at today?
George Hershman: Well, actually, we’re really excited about the market. We’re seeing huge opportunities across all of our services, whether that’s our EPC business, our O&M business, our HV business. I mean, the power demand is real and we’re seeing growth in every market across the country. So having been in this market for 18 plus years and seeing that demand is growing so fast and the need for our product and solar and storage has never been higher. So I’m pretty excited about it.
Shayle Kann: All right. So boom times for solar and storage and all the associated services that you guys provide along with that, you mentioned you’ve been in the market for 18 years. I’m actually interested to hear, is this a cyclical business? You think of construction in the context of housing and things like that as being very cyclical and you go through periods of boom and bust and then the construction business kind of follows along suit with that. How have you seen over the 18 years, how has this market evolved? Is it similar boom and bust periods or is it more, I don’t know, odd than that?
George Hershman: Yeah, I think it’s a little bit different. I mean, I spent my first two thirds of my career in commercial construction, so I saw that cycle kind of every seven years. We don’t really see that the same. And renewables has grown consistently since we started the business in 2008 and we started in small commercial and then that moved into utility. And I laugh now when I think about a utility project was like 10 or 15 megawatts.
But we’ve just seen a very consistent growth period. While we’ve seen, I think, small dips in the market in 17 and 18, there was a little bit of a development hangover from the original, what the ITC cliff we thought was there coming in 16. But since then, it’s been kind of a slow march. If you look over a 24 month period, and I like to measure much more than kind of year to year, but if you look over any kind of 24 month period, there’s been a gradual growth. But now, we’re seeing significant growth because of demand. And I think that’s important, right? Is when you think about how we all built our business, we built it in an energy market that was fairly flat. A lot of that was replacement of aging assets, but not real load growth. And now we’re seeing significant load growth that the only way to meet it is with solar and storage, right?
When you think about what is the fastest to deploy and lowest cost of energy, it really puts you back at solar and storage. And so I think that’s what we’re really seeing consistently. And so there’s not the same kind of bust kind of boom and bust periods. I think what we’ve tried to do with our business a bit is look at how do we smooth out some of the transactional nature of EPC business. And that’s why our kind of life cycle approach, thinking about the EPC business and the O&M business coupled together and providing those services for projects over a entire 35 year plus life cycle. And so while we might see some softening in EPC at different moments in time, we’ve got this reoccurring revenue base that we manage over 20 gigawatts of projects across the country. So we have this large kind of recurring revenue base that smooths out our business maybe different than other ENC companies and EPCs in the country.
Shayle Kann: I have always wondered how you manage that. I imagine in an ideal world, so you have a bunch of … Let’s just focus on the EPC side for a second. I appreciate you. You have the O&M business, which is recurring and more consistent, but if you just look at the EPC side, you’ve got a bunch of full-time employees and then you also work with, I’m sure, a million contractors and subcontractors on any given project, but you’ve got some fixed cost base and that fixed cost base, I think you could map it to a maximum number of megawatts that you could be constructing at any given time, I assume. And it could flex up and down because you could add more or less contractors, but these projects have gotten so big now. You guys are probably, I don’t know what you’re building right now, but I’m guessing you’re probably building multiple projects that are multiple hundreds of megawatts, maybe some projects that are approaching a gigawatt or more.
And so they’re these big lumpy projects that last a certain amount of time, some number of months from the beginning of construction to COD. Are you basically in a constant game of trying to put the puzzle pieces together such that you have a flat number of megawatts in construction over time? Is it like a Tetris game like that or does it look something different from that?
George Hershman: Well, it’s actually changes over time. As we’re starting to see these projects get larger, we’re actually getting operational leverage. Because you think about it, if we were building 20 projects a year, say, and all those 20 projects were 20 megawatt projects, I’d still need a team to manage each one of those projects. Now those projects are averaging over 300 megawatts and up to a gigawatt in scale. So for a … I don’t need 10 teams to manage a gigawatt project versus a hundred megawatt project, right? So I actually get leverage in the system. And so a project team, and maybe with some additional management resources, can do a project that’s three or 400 megawatts versus that same project team that would do something traditionally a few years ago at a much smaller scale. So the megawatts are growing, but the people that are managing them are not growing.
So there’s a lot of leverage in the business as you start to look at these projects at scale. So if we had 20 teams building 20 gigawatts, I mean 20 gigawatt projects, we could talk about a 20 gigawatt run rate, right? I mean, yes, I would have to add more skilled labor resources, more field labor resources, but those are largely trainable skills and people that I can recruit. So I think that’s why we’re seeing the business just continue to grow over time and be able to take on the billions of dollars in revenue that we’re starting to see from a … Yes, we’re growing. I mean, we have 2,600 employees, so we started this business with one or two. And so now, yes, the business has to grow with scale, but it doesn’t have to grow a hundred percent even though we’re doing that much more revenue. And so I think that’s the difference that we’re starting to see.
Shayle Kann: You mentioned something there that I think cuts a little bit against conventional wisdom right now, which is you said, look, if we had to expand and do 20 gigawatt runway, we’d need more skilled laborers, but that’s something we can train. I think we hear a lot about a labor shortage and not specific just to solar and storage construction, for example, but now with this rising data center construction boom, we hear it there as well, and a lot of those people have the same skill sets that need to be trained, electricians and so on. How much of a constraint do you see the labor pool as? If you did need to go to 20 gigawatt run rate, how hard would it be to recruit enough skilled laborers?
George Hershman: Well, I think you have to think about our business a little bit different than building a data center and the electricians and the skillset necessary for a data center. If you think about a solar project, probably 85 to 90% of that is really labor and mechanical labor, right? We’re putting posts in the ground, we’re bolting together racking systems and we’re installing modules, and then 10 to 15% of that is really skilled electricians that are doing terminations and splicing and the kind of highly technical skill that you need to do those type of activities, where in a data center that’s happening every square inch of a project, you’re doing those kind of terminations and electrical work. So while we have highly skilled resources in our HV, high voltage groups and doing substation work and where that actual highly skilled resource is needed, a lot of what we’re doing is bringing in local resources and training them to do the mechanical install portion.
Now, that’s not to say that we’re not looking at optimization. So we’re looking at every robotic system that’s out there, we’re doing testing on every one of our sites has some form of automation or robotics that we’re trying to implement because we’re looking at how do we optimize crews because labor is expensive. And so even if we could get it, we want to reduce it or optimize it so that we can deliver at the lowest cost and continue to look at ways to minimize the cost and drive up the efficiencies of our projects. Because I think the questions we get asked more is not about how can you build a project of this scale, but how do you build a project of this scale faster because—
Shayle Kann: I was going to say, it seems like the speed actually, when it comes to automation, you get the benefit of a lower labor pool. If you’re constrained, you get the benefit of it maybe being cheaper. But right now, it feels like the killer app for automation is speed. And I’m curious how you think about that in terms of what is the length of time that it normally takes you with a fully manual labor world to build a 300 megawatt project or whatever size you want, and then how much might you be able to speed that up as you look at these automation solutions?
George Hershman: Well, not every project is the same, so it’s really hard to come up with a … 300 megawatts takes you 12 months full stop, right? In certain areas, you could do it in 12 months, certain areas you’re going to do it in 16 months, right? Because it’s harder to build in the northeast than it’s hard than in West Texas. So we’re trying to find out how do we just cut 15% or 10% out of a total build, right? So we’re looking at more of how do we build in that type of efficiency. And look, I always tell people, if I could transplant a built site, like pull it off the shelf and put it on the ground, I’d be sold out, right? So our customers just want to say, how much faster can you build them? And so that’s what we’re looking at. I think in certain regions, we’re going to gain 20% efficiency in some areas just because of weather and conditions and the buildable time that you have in certain regions, it’s just going to take longer to build.
And so I think it’s for our teams, and we have a team fully dedicated on how do we build it more lean, with more innovation and technology within the process, because I think we look at our business much more akin to manufacturing than we look at it as construction, right? You think about, we’re trying to do the same activity a million times, and how do we do that just slightly more efficient? And that’s going to be through some part of human resources and some part of automation and robotics. And those two are going to have to work together to optimize and continue to drive out time to market.
Shayle Kann: I’m curious what you would say the rate limiter on your growth is right now. If I told you you need to … I mean, we talked about getting to 20 gigawatts, but let’s just say as a theoretical exercise, I said Solve needs to be at a 50 gigawatt a year run rate as fast as possible. What would be the hardest thing to scale up?
George Hershman: I think logistics, right? I think that when you think about these projects and the scale of them, it really becomes a logistics officer type of business, right? When you think about … They always said, logistics officers won wars, right? And because just moving the amount of equipment that we have to move and the amount of trucks … So when you think about these projects that are, how do you build … And a question get asked of me, how do you build two gigawatts in a year on a single site when you think about these massive data centers? And I said, tell me how I’m going to move that many trucks, how I’m going to … Assuming I can get all the people, right? Because we’ll attract people, we’ll attract, we’ll get … How am I going to continue to supply that many people putting that amount of work in place, right?
The amount of trucks that we have to move becomes the gating activity, not the amount of people, right? Because how are you going to park people? How are you going to bust them in? How are you going to … I mean, it’s all becomes a logistics game more than a construction activity. Once I get them to this location, I can put posts in the ground, I can bolt stuff together. It’s just a matter of doing all those things at the scale that we’re talking about. And look, the great thing about our business is we have been mission focused on this since the beginning, right? We have specialized in solar for two decades and storage came in the last five years, and we’ve been focusing on how to do this more efficiently with everything that we do. And I think that is unique in when you think about, there’s a lot of businesses out there that are diversified and they’re thinking about how do I build a interstate transmission line and how do I build a pipeline and how do I build all these other things?
And our business has always been about how do we build a utility scale solar project and how do we build it the most efficient possible? And I think that’s why we’re seeing a lot of these small gains that are necessary to meet the kind of business that we’re in today. And I think that when we start to see these projects grow, I think that we’re going to have a lot of opportunity to see these pieces of technology start to realize, right? I mean, I think we’re not … I mean, on a one-for-one basis, I can still use human capital and labor to produce faster than robotics that are out there in the market, but could I use robotics on a third shift or could I do these things that really optimize speed? I think that’s really what we’re going to see.
Shayle Kann: You mentioned the logistics orchestration. What I’ve seen that has always struck me at these big sites is the degree to which there’s this orchestration of equipment delivery as well and the timing of all the equipment delivery. And you see these situations where a project is chugging along, but it gets gummed up because this lay down yard doesn’t have the torque tubes delivered at the right time or whatever it might be. How much … We’ve seen these kind of crazy swings in supply chains over the past few years, broadly, as a result of tariffs and all sorts of things. How much has that affected your ability to appropriately resource and predict timing of individual construction sites? Is it a well oiled machine or is every site kind of like a tiny bit of chaos that has to get abstracted away at the site level?
George Hershman: Well, I think the biggest impacts we’ve had to our business over the years has been supply chain disruption, right? I mean, we all lived through 22 and 23 when we saw massive disruption in the panel market and solar panels were stuck offshore, they were stuck at the border. We had projects that were built with racking systems and no panels, and that disruption is massive to our business because no manufacturer can build their product without supply chain. And so you’re right, making sure that our crews are adequately resourced because without the product there, it’s, one, is detrimental to the schedule, but it’s a huge amount of cost because we have people that are not as efficient because they’re waiting for product or we’re demobilizing and remobilizing and it’s expensive because you lose people, you have to retrain them, do things. So that’s why I said logistics is one of the biggest challenges and one of the biggest opportunities.
If we can keep the machine fed, we can build at high rates and speed, but if we don’t have a piece of material or a torque tube or something that we need at a certain time, then the whole system gets slowed down and that means we’re catching up through the rest of the project because we have to keep these cycles and process moving. And so, we have expended a lot of resources and a lot of time in developing great relationships with our vendor partners and recognizing what we need to keep the supply chain moving. We’re able to invest early on long lead time activities and items so that we make sure that product is delivered ahead of schedule because there is no just in time delivery on these projects. If you do just in time delivery, like that was the whole mantra of my commercial life when I was building commercial construction is that you were on small constrained sites and you would deliver just in time because you couldn’t move material around and you needed it there when you needed … The day before you needed to install it.
If we do that on these projects, we’re late because the momentum is going so fast that we need to deliver early and make sure that where we’re delivering it is in the general work location so that we’re not double handling and having a lot of expense, but I would rather deliver onsite and double handle versus the concern about not having material when I needed it. So that’s really what we’re doing a lot of these days is spending a lot of time on thinking about what can we buy, what type of early, what type of strategic procurement can we do so that we have … If we use 80% of the same cable on every project, then just buy out months, if not a year ahead. And so we make sure that at least the non-specialty items are there long before we need them. And then we’re only buying the 20% of cable that is very unique to a specific job.
And so, I mean, those are things that we’re thinking about all the time as we scale the business. And we have a lot more resources today in procurement, in pre-construction, and things that get us set up for success versus a prototype construction project where a project team is dedicated to all the pre-construction and all the procurement. We have centralized pre-construction. We have centralized procurement. We are looking at our whole fleet of projects that we’re building and moving equipment and resources around as if it was all one cohesive project.
Shayle Kann: This is, I guess, a good segue to another thing I was curious about, which is we’ve talked about robotics and that’s one instantiation of AI, of the new AI wave, but I’m curious, what other areas, if any, in your business you see either real movement as a result of AI or potential? So I could imagine system design, I could imagine logistics and supply chain. What are the places where you see AI penetrating your business apart from robotics?
George Hershman: Yeah, no, I think that’s the interesting part of our business. And I talked to a lot of people and I seem to always get the question, how is AI affecting your business? And we’re actually on both sides of it, right? AI is driving our business, the reason why … And AI, and then we are using the same technology that’s driving our business to try to optimize our business. So we’re looking at everything from what is the best plant optimization, like how do you lay out the logistics plan for a project? So if you can imagine doing simulation models on just plant layout, where do you park people to get them to the work site the most efficient? Because I’ve challenged our teams to think about how do we make every person on the site 15% more productive, right? Or find 15 minutes of more productivity out of every person by the … Because if you think about projects that are measured in square miles, where you park them, where you park employees and how you bus them, because everybody checks in, so you’re on the clock, you get into some sort of transportation, you go out to your work site, and then 20 minutes before the end of your shift, you get back on that bus and come back.
How do we figure out how to shorten those periods of time where people are working in their eight hour day, but they’re not actually productive because of just internal logistics, moving people around, where is their break area, how close are the restrooms to their work area? I mean, all those things we’re looking at and using AI to run those type of simulations so that we don’t have to build it and then determine that we would have done it differently. We can run, we can put plans in, we can put in mobilization plans and do a lot of simulation around how best to move equipment.
What if you park them in the north corner versus the south corner? How close is their work area? What if you do two lay down areas and not one lay down area? How many moves will that eliminate in or minimize during the course of a project? And all of those can be done through simulations now where we used to have to kind of measure them in real time. And so I think that’s where a lot of our technology team is focusing because we collect so much data on a given day of production, how do we optimize it?
Shayle Kann: Do you find that for things like that, you have to custom build your own tool to do that kind of a simulation? Obviously you have proprietary data that you want to leverage, but can you use Claude code or whatever to go build something like that or are you guys building your own like bespoke vertical applications?
George Hershman: No, I mean, I think they’re a little bit of both, right? We have a purpose built internal platform that collects all of our performance data. So we built a program, we started about 12 years ago now and built a program called Sunscreen that collects all of our project data so it gets collected in real time. We can take that data and we can run it through more conventional Claude platforms and other things that our team is using to be able to extract the data and really tell us what we’re seeing, just run these simulations, but it’s a bit of a mix, right? We use some commercially available tools and some Some of the stuff that we do internally, and this is where having an internal software dev team has been really helpful. Because of sunscreen and because of our Vitals platform, which is an internal platform that we built to manage plants, we collect so much data and we can run it through those platforms and then we can run AI kind of overlays on top of them.
I think that has allowed us to be a real kind of a technology leader in this space.
Shayle Kann: All right, George, I’m going to let you go. Thank you so much for your time. Great. Super interesting.
George Hershman: I appreciate it. We’ll talk to you soon. Thank you. Shayle Kann: George Hirschman is the CEO of SOLV Energy. This show is a production of Latitude Media. You can head over to latitudemedia.com for links to today’s topics. This episode is produced by Max Savage Levenson, mixing and theme song by Sean Marquand. Stephen Lacey is our executive editor. I’m Shayle Kann and this is Catalyst.
