Good morning and welcome to our second session of our third annual Fiber to the Home Symposium . My name is Greg Williams. I cover the cable, wireless, and telco space here at TD Cowen. This will be a fireside chat. Delighted to have along Steve Nielsen, the President and CEO of Dycom Industries. It'll be a 45-minute discussion. We'll leave some time for Q&A. If you have any questions, feel free to enter it in the question box. I have a screen over here that sees the questions. Of course, we'll keep it anonymous. And without further ado, Steve, thanks for joining us. Appreciate it.
Oh, Greg, thanks for inviting us. Just before we get started, I wanted to remind everybody that during our discussion, I may make some forward-looking statements which are subject to risks and uncertainties, which we've more fully outlined in our SEC filings, which we recommend for everybody's reading. With that, good morning and glad to be here.
Yep, glad to have you. Let's start with a brief overview of your company. For those in the audience who might not be that familiar, what does your company do? What is your role in the fiber to the home deployment process? Who are some of your customers and what's keeping you busy?
Yeah, so we just completed. We're a January year-end, so we just completed our fiscal year just under $4.2 billion in revenue. We have just under 16,000 employees. We work strictly in the U.S., so have no international presence. And we're the folks that engineer, construct, and maintain telecommunications networks, whether that be for wireline telcos, wireline cablecos , wireless companies, increasingly in rural America, some electric co-ops. So just anybody who's out there who owns, operates, and is extending or building a network, we're heavily involved. Not a league table business, but I think we're generally acknowledged to be the largest provider of wireline services in the country.
When you mention wireline services, we just had Frontier moments ago on the call, Verizon, of course, AT&T. They're pretty heavily unionized in doing their builds. Relatively speaking, I'm just curious, when do they go and choose Dycom versus keep it in-house? What are the gating factors?
Well, the good news is most of those decisions were made decades ago, Greg. So if you think about the modern-day contracting industry, I don't know if it goes back to the founding of AT&T, but there's always been a role, particularly for heavy construction in urban and suburban America, that's always been contracted for the most part. And then aerial and splicing services and other technical services as a function of kind of what union agreement was signed when. But I always give the example of Knoxville, Tennessee, Greg. We have provided the outsourced services to whoever operated that phone company since 1954. So this will be 70 years. They tried somebody else for 3 years in the 1980s, and we've been back ever since.
So I think when it comes to those places where they have a choice and it wasn't made decades ago, typically the work is outsourced. So if we think about all the emerging overbuilders, that is predominantly outsourced. It's tough enough coming up with the right capital and a higher cost of capital environment to deploy the network. Deploying capital to build in-house capacity has just generally not been where people have been focused.
Right. And how much of fiber is part of your overall business, that Dycom?
Yeah, so the maintenance and operations part of the business, which includes routine CapEx and routine fiber deployment, is the majority of the business. And that's the great thing about our business. Not only do we benefit when our customers are deploying new technologies, whether that be HFC or fiber or 10, 15 years ago, copper, but we also benefit because we're around to help them operate that network. So in large parts of the country, anything that touches the underground portion of a telco's network, if they need to get to it, they're calling us or somebody like us.
Right. And where do you see Dycom in this fiber narrative? Is it year-over-year when you think about 2024 or 2025 and then even 2025 and beyond? How do you see fiber shaking out in terms of Dycom's builds?
Well, one of the things I can do, and if we looked at, we're a little bit off on fiscal year, but if we look at calendar year 2022 and 2023, we grew about $950 million organically, a little more than that. We did a couple of small acquisitions. So we've seen lots of robust growth opportunities. We participate in all parts of the country, so rural, urban, and suburban. And we see opportunities in every one of those subsegments of the market.
Right. I just want to switch gears a little bit to BEAD. I know we have a panel at 1:00 P.M. with our Washington policy analyst, Paul Gallant. But you're in the trenches with BEAD. It'll be a big part of your business, I presume. Just curious to hear your thoughts. What are the further hurdles and challenges through the balance of the year with BEAD? And maybe when will we see the shovels in the ground?
Yeah, look, we're following it closely like everybody else. We have folks assigned to that task. I think what I would say is that BEAD was structured in a way different than the 2009 stimulus program that we also were a substantial beneficiary of insofar as the money goes to individual state broadband offices who in turn make subgrants. So right now, as I understand it, the states are working through with the NTIA, so that agency is part of the Commerce Department. And they are working through kind of the details of what conditions will come with the receipt of money by subgrantees. And right now, there's a little bit of work to be done on that topic.
I think historically, when you have a big pool of public capital that's available, that can do an awful lot of good for a lot of rural America, that'll get worked out. But they're still working through some things right now on a state-by-state basis, which I think is actually a healthy process. But better to work through timing or conditions now than later. I think the best thinking right now is that we'll see subgrantees beginning to be designated in the fourth quarter. And we think there's activity going into 2025 and certainly accelerating from there.
Right. And you mentioned RDOF, and there's been reports that saying things like 30% of the RDOF builders were in default. There was a report with Altice recently. I'm curious, how do you think that BEAD sort of shakes out? It seems to me that presumes that some of the larger players would win, the AT&Ts and the charters, rather than smaller players. And I'm curious to hear your thoughts.
Well, I think we've read some industry research that talks about how much of the BEAD passings are in each of the incumbents. I think I saw something publicly where, for example, Brightspeed, a good client of ours, had about 1 million homes passed, more or less, in their footprint. And I think anytime those kind of dollars flow to incumbents that already have central offices, so they have places to put fiber optic equipment, they have conduits to the extent those are needed that are already in place. And probably most importantly, they have pole attachment agreements and access to their existing strand and cable to lash fiber to. That I think that is the most likely and easiest path in those areas where those incumbents have a footprint. Where they don't, I think obviously you mentioned Charter.
They've had a very aggressive RDOF program as well as state program. Interestingly, a couple or three months ago, they actually said that based on those programs, they had identified another 300,000 homes that now were deemed economic without subsidy because they were adjacent to areas where they were receiving subsidy. We're really encouraged, as again, we've been through federal programs before, got our fair share, and see no reason why that won't happen again.
It's beyond BEAD too, right? I mean, there's other state programs that are really adding up for you guys.
There are state programs. There's RDOF. There are lots of programs that are in the marketplace. And it's interesting. Sometimes we'll get a build. And over the length of the build, there may be three different funding sources. It may be the owner's own capital. It may be a state program. And it may be an RDOF all in one actual project that we have secured.
Right. Wanted to shift gears a little bit and talk about open access because I think you build for a couple of the open access players as well. What are your thoughts on the fiber to the home open access model? Last panel, there was a little bit of mixed messaging, some dismissive, same thing. It could work in the right spots and the right places. But do you think it could massively disrupt the industry for your customers? Or do you believe there could be meaningful traction in the U.S.? And why or why not?
Yeah. I think of open access and new business models as indicators that there's a lot of capital looking for ways to provide high-capacity connectivity to America. And that it's really a healthy sign that people are examining every available alternative to do that. I would defer to the folks on your prior panel as they think through it. They certainly have a lot more experience with it or have thought about it more than we have. I would just say that we're always encouraged when new capital sources are applied to an industry that has so much opportunity as the one that we participate in right now. So I think new business models, whether they eventually play out or not, are just a sign of the health and the interest in the industry right now.
Right. Fair point. And the business model doesn't really change for you at all, right? You're building regardless of the model on the wholesale retail side.
Yeah. We're typically, we've never figured out a way that any of these models work without physical construction. You got a place to plant. And while there are different approaches to contracting, they're pretty much, we get paid by the foot or we get paid by the mile. So it doesn't really make any difference what the business model is on the back end.
You brought up a good point there. How do you sort of price out your-do you price it by neighborhoods? You mentioned you price by foot or mile. And yeah, maybe you can discuss how it goes from bookings to installs to revenue.
Sure. So one of the great things about our business, and we disclose this, is if you look at our multi-year master service agreements and other long-term contracts, they're typically about, I think, somewhere in the neighborhood of 80% of the business, sometimes a little bit more, could be approaching 90. And so most of those agreements are described geographically, so they cover a specific geographic area because geography drives costs. So that's an important factor. The other thing is they typically have hundreds of individual tasks that describe the work that were issued one work order at a time. And so we generally get paid by the foot or by the pole or by the splice or burn of fiber.
It means that for us, it puts a premium on being really good at managing the back office elements of a build because not only do you have to get it built, you got to know exactly how you're going to invoice it and get paid.
Right. And on the cost side, we sort of touched on it last panel. It's 2022. We were in sort of a crisis mode on supply and labor. 2023, we managed it a little better and a lot better. And then 2024, even better still. Where are you seeing the labor and costs shake out in 2024 and then beyond? And if I can add an additional question, then when I say beyond, with BEAD coming in, will that be a crisis because we're chasing that same finite labor pool in 2025?
Yeah. So if 2022 is your baseline for any analysis, I'm not sure what the historically comparable periods are because clearly, as we're still dealing, I think the peak in terms of impact on our business was in January of 2022 with Omicron. I mean, we had a lot of people that were out of work. And so I think people, we've kind of all forgotten that there were lingering effects of the pandemic even into 2022 just directly in terms of impact on people and their lives and their health. So that aside, what I would say is in every large program that we've been involved with over decades, because these programs make such strategic sense to customers and to the industry, it seems like everybody has very high expectations for year one. And I coined a phrase a long time ago that sometimes Wall Street over-anticipates and underappreciates.
I think sometimes that's true of supply chain, right? It's difficult to grow. In our calendar 2022, we grew almost $700 million. You can say that growth was tough for the industry, but we certainly grew lots and found the resources to do that. As time has passed, I think not only us, but our peers as well as our customers have gotten much better as time has gone by to manage that growth. I'm encouraged that I wouldn't call 2022 a crisis, but it sure was a challenging year. I think that those challenges have diminished as time's passed. I mean, these are large, complicated programs. You deal with hundreds, if not thousands, of individual permitting agencies. That creates bottlenecks. In 2022, there certainly were material bottlenecks that were pandemic-related.
And so that only makes the, puts another level of complexity on an already complex initiative. But I think over time, I'm always encouraged as we work with our customers that we get better, they get better, and more work gets done. If we look ahead to BEAD, we see BEAD as an incremental driver of demand. We've been benefiting from public capital under the CARES Act, RDOF. There's something north of $20 billion of state capital, unrelated to any of the federal programs. And so we see that as continuing a trend, but with one caveat. What I'd say is this is a dynamic industry. The amount of capital that's attracting both public and private is really stunning. When you start adding things up and you're in the hundreds of millions, it's a big number. And so we're all going to have to work hard to make that happen.
But I think it will get done.
Right. And with that capital coming in, I mean, we've got cost to build. I'm curious to hear what your cost to build generally is. I know it's very tough. It's aerial versus underground. And it's the soil. And it's the regulatory and permitting challenges you mentioned. But what is the cost to build a home passed these days, generally? Or maybe you can just answer it in a better way than I'm asking in terms of where it's going.
Yeah. I mean, look, what I would say is anytime you have a bunch of programs that are starting off new, everybody works through a model of what the mix of aerial and buried and then multi-dwelling unit construction is. It takes time. And those can change over time, right? So it takes time often to gain access to MDUs, for example. And so there are MDUs you might have wanted to do early in a program, but you can't gain access to them till later in a program. So you have a model where the mix are, and the mix issues are significant. And again, all your caveats included, but buried is much more expensive than aerial. And so really what happens is not necessarily unit cost, although we've got to earn good returns so we can attract the resources to grow for our customers.
What we really see is more tightening of the models on that mix. That may cause some cost to change over time. The unit costs have been reasonable over time.
They've been reasonable over time. I think there's a fear on Wall Street that some of the low-hanging fruit is sort of done. I guess I would consider that aerial maybe. Is the mix to buried probably more likely, or is there still a land grab? There's plenty to go. I'm just trying to see if the costs will rise, not necessarily because of labor costs, but because just the mix shift.
Well, there's always going to be mix shift over time. Some of it's just random based on geography. Sometimes it's a function of rural builds, which can be much more aerial or can have cheaper underground construction methods just because that's available in rural America where it's not in urban and suburban America. But on the other hand, there are also cost advantages, right? Setting up the supply chains and the logistics to start one of these million, 2 million, 3 million home builds from scratch, there's lots of expenses there that you begin to amortize as you get deeper into the programs. There's an amount of feeder fiber you have to get out there before you can start doing the distribution fiber that passes homes. And again, once you build the ring, it's built. If you start adding subdivisions to it, you're amortizing that cost.
It's not always clear. I always take great solace in the fact that, and I saw in your last panel, you had a representative from Verizon. March 11th marked our 20-year anniversary of working on Fios somewhere, not always at a high level, but we always were working somewhere in America on that program. The fact that they're continuing to do that today and pleased with the returns tells me that there's plenty of runway here for good returning work for others.
It's not just aerial and underground, and then we're. Now we're going more rural, of course, with BEAD. Maybe you can help contextualize the challenges. Obviously, a truck roll will be 90 miles away and things like that. You're in the trenches, pun intended. Maybe you can talk about the rural builds going forward and the challenges there.
Well, it's interesting. We probably have as much footprint as anybody else that I can think of in terms of serving rural America through these master service agreements. So we have lots of master service agreements, really, in every part of the country for cable and telephone companies where we're there. And where spending has not been all that robust, probably since first-generation DSL or first-generation deployments for the cable operators. And so we're kind of encouraged that there's more work moving to rural because, in large part, we've got lots of not-as-utilized facilities as we'd like in that part of the country.
I mean, one thing for certain, when you get big builds in rural America, and we've been building for electrical cooperatives, which almost by definition are rural America now for over 10 years or almost 10 years, is you're not going to have enough local labor available to build what that individual county or group of counties is going to build. If you get a project in the middle of Nebraska for 5,000 miles, you're not going to have local labor. You're going to have to bring that labor in. And that's just part of the model because when you're done, they'll move on to the next project. In terms of truck rolls, it's very interesting.
I had a discussion with a client who was considering just going ahead and pre-positioning all of the drop connections from the mainline cable into the homes because they really didn't want to do an on-demand dispatch and have to come back if it's a two-hour ride. So I think business models will evolve in rural. That may be a little bit different than other parts of the country. But it's still a pretty ingenious country. If there's an opportunity to create a return, people tend to figure it out.
Right. When there's money and returns, they go for it and find a scalable model with BEAD coming up. Maybe I want to discuss the cost to connect as well. How have you seen the cost to connect a home fluctuate up or down? Labor's up, the bucket truck and the person in the bucket truck. And then, of course, the equipment, the ONT on the side of the house, that's down. Maybe talk about what's the mix of a cost to connect and how is that shaking out?
Yeah. We don't supply the ONTs generally, but I'm told that from 20 years ago, they've dropped from $400 or $500 to under $100. And so that's a significant cost to connect that's had great reductions in cost. And I would tell you that the biggest challenge in our cost structure, of course, is it's a 4% economy to attract, 4% unemployment economy. If you're going to attract new labor, it's got to be rewarding for them. They got to come from someplace else. We have training programs to grow folks. And actually, the work that you're talking about in drop is an area where we are growing capacity right now because there's much more activity. One of the interesting things about the big builds that people forget, although I did hear discussion about penetration rates over time on the last panel, is it takes time to penetrate these networks.
So to the extent that we're involved in drop placement and operations after we place the network, that can have a 4-5-year tail as customers get to terminal penetration. So everybody likes to think about the construction, but there's a lot to do afterwards.
Right. You sit in an interesting position because you help build fiber, but you also help with the cable network upgrades as well. I'm just curious to hear your thoughts on the meaningful cable upgrades. We came back from the Cable Next-Gen conference last week. Right now, it seems like it's DOCSIS 3.1 Plus or Ultra or whatever you want to call the 3.1, high splits, mid splits. Then, of course, they're going to be doing DAA to get to DOCSIS 4.0. How much of it is your business? How do you see the shakeup between the two different kinds of customers you sit with between cable network upgrades and fiber to the home?
Yeah. We do provide lots of the technical services, both from a design and engineering perspective, as well as the actual technical services to upgrade high splits and mid splits, as well as change out nodes and do that activity. And I think, look, the cable operators are in an enviable position. They invested a lot of money in the '90s and through the early 2000s to upgrade their HFC plant. They're going through another cycle again. We're pleased that we can help them do that. We don't tend to take sides. We're just happy that we're able to work for everybody. And it's always interesting to see how big themes intersect with individual company strategies. And everybody solves it a little bit different. And that's what makes the business so interesting.
Right. And longer term, where do you see the fiber builds sort of playing out? I mean, how many more years of this sort of frenzy of 8 million homes do you think we have? Or maybe another way, if you put your macro hat on, I mean, how many homes will be served with fiber, from you see, the costs pretty closely?
We've talked about this and also our imprint. We think ultimately, we don't have a magic crystal ball. But we think ultimately, 75%-80% of the country gets two high-capacity networks, one from a telco operator or somebody else using a fiber network to provide that service, and then the cable operators. Again, cable operators in the greenfield world and the rural world, they're deploying lots of fiber. I think there's focus on all the DOCSIS and 4.0 and all the other things. But they're big fiber installers today in addition to that. So I think that 75% or 80% gets you north of 100 million homes. We're probably in the low to mid-60s right now. So we got a long way to go.
And then when you think about it, you have rural of, let's call it, 30-35 million or some combination of rural and semi-rural. And I don't know that they all get built. Certainly, there are some where satellite's going to make sense. But the thing to keep in mind is as you do those homes that are predominantly rural, because the density is so much lower, the cost per home is much higher. So even if you think about it, if rule of thumb, the industry uses $1,000-$1,200 homes passed for capital investment in urban and suburban America. If you do 20 million homes in rural America at $3,000-$5,000, you got a lot of effort that has to be put forth.
So sometimes we get focused, Greg, a little bit on homes, like if 8 become 7, but 3 of those 7 are rural, it'll be like 11 or 12 under the current basis. So lots of work to do.
Well, that's what's also interesting. And curious to hear your thoughts is I feel like there might be a donut, if you will, because you're going to have the rural areas from BEAD get built, and then the urban and suburban areas can get built just from the capital markets alone. But then you've got that kind of fringe area. And curious to hear how folks are doing that.
Yeah. Look, it's interesting. We work with a large overbuilder that came into a metropolitan market about 10 or 12 years ago. And they were very disciplined about how they thought about the build, right? So here are neighborhoods we're going to build first, here are neighborhoods we might build second, and here are neighborhoods that we're probably never going to get there, right? That's your concern. And what they found was, and I think you've had people on your panel that have talked about this, that there were actually efficiencies that could be generated. So when they built all the neighborhoods that really made sense, all of a sudden, the economics on the next tier got better because there was all that fixed infrastructure to serve the first cohort that was available to be amortized over the second cohort.
Son of a gun, they got through the second tier, and all of a sudden, the things that maybe never made sense started making sense. So when you add this public capital element, if you have to take transport from where you are to where you need to be, and you can get support for that from some public entity, then all of a sudden, all those areas in between may make more sense. It's going to take a while. These are big networks. They're building for a generation. But I'm ultimately optimistic that high-capacity networks are so integral to the way we live today. And that's not changing. Every generation is going to become more and more focused on digital applications that use lots of bandwidth. And I think this will get built. Always hard to say over how long a period, but it'll get built.
Yeah. You bring up a good point I don't think about much, is there's pros and cons to the business case evolving. Cost of capital going higher, labor and maintenance, or labor and the building equipment might be a little higher. But then ARP is good, and penetration's getting better. But you just mentioned you already built out hundreds of thousands or even millions of homes. And you've got those sea legs under you, if you will. And you can get out to homes you otherwise maybe not just from the learning curve.
Well, I just think it's generational, Greg. If you take kind of the 100,000-foot view or a long-term perspective, there were generations that couldn't live without a telephone. You had to have the ability to call somebody. And we had a generation that couldn't conceive of entertaining themselves without TV. Who today thinks that this generation or the next can live without high-capacity data connections? I just think it's that fundamental. And I think that was, with all the dislocation and all the pain and misery of the pandemic, I think that just made so clear the value of what our customers provide the country and, by extension, what we provide our customers. And I just don't see that changing. I just think that's a generational shift that will continue.
Yeah. When I move into a new home, I probably land the broadband before the heat and the water, right?
Exactly.
I didn't get a chance to ask this on another panel, but I'd like to ask you is about AI. I mean, it's obviously taken over, at least in the data center world, etc. But in the telco space, it's a little more elusive. So maybe you can help us with your thoughts on how do you think AI will drive both external opportunities and then maybe just in-house efficiencies, etc.
Sure. So we'll start with external. And look, it's new, and it's topical. We couldn't get through something like this without an AI discussion. And some of this is anecdotal, and some of it's just my speculation. But it's amazing. Early in my career, we built all these point-to-point long-distance networks. And when the dot-coms imploded 20+ years ago, everybody's view was, "There'll never be an application that uses up all the dark fiber or dark conduit capacity in this country," right?
And yet now, with the amount of data center capacity that allegedly is going to be required to do AI, we're seeing and hearing about opportunities where that original 1980s, late to mid- to late '80s through early 2000s fiber networks are going to need enhancement and reinforcement and where data centers have to be more diversified based on the electricity that they require that requires really high capacity fiber connection. So it's just another great example of every time in this industry when I've been told that some technology has eliminated the need for services forever, if you wait about 10-20 years, there's always something that comes along. And in this case, could very well be AI. And so I think that's great for that part of the network, to the extent that AI really takes off and latency is an issue.
This was something that we talked about a lot with autonomous vehicles, maybe not as much recently. But you could clearly see a premium placed on more small-cell activity so that you get really ubiquitous, very high capacity, very low latency wireless connectivity. And that's what we're looking for in our business, right? So we have, call it 10-12,000 technicians out there every day. Several thousand of them work as a single technician work unit or crew. And the more ubiquitously we can access high-capacity connections to route them, to help them be safe, to ensure that they do the work the way we and our customer expect it to be done in real time, I think those are all great AI applications that will require really high-capacity connections.
I mean, if you think about, wouldn't it be great if instead of trying to either have a supervisor communicate the right technical method or the right craft method at a point in time, that you called up a digital twin and you watched some hologram actually complete the task, right?
Right.
That sounds far-fetched. I think you could probably do it today in a lab environment. Wouldn't that be great if we knew exactly how people were completing this particular fiber splice, that the test results were all not reported on but just sensed because we had everybody in a network? I think there's great opportunities for productivity going forward, which is why I've been hearing for a long time, Greg, that the cost of something means that we'll never build it. And we're just in this great industry where what we build for our customers actually helps us be more productive, which means in turn that they can spend more as we get more done with our folks.
Right. It's like a virtuous circle, I guess. You help build these customers.
I mean, we have a device where we communicate with all of our crews. We have thousands and thousands of iPads and iPhones out there. It gives us ways to manage our workforce that 10 years ago, we might have thought about, but we would have never been able to do.
Right. I look forward to my digital twin fixing my garbage disposal. But joking aside, it's funny that you mentioned low latency too because you asked the cable folks today. Latency is really a marketing thing, right? Because if the other guy offers it now, we offer it. And since we're already going to DOCSIS 3.1 or plus, latency now is just sort of a me too sort of thing. But then latency could actually be something that could be a competitive advantage.
Well, you asked me to speculate. And look, I think the cable plant will always evolve and always be competitive. They have their fearsome competitors. So I never would gainsay what they're doing. But I'll tell you what, right now, it might be a little odd to think about it. But if I could have a digital twin of every one of our technicians that's doing a high-risk, very critical task, and we could supervise them digitally, that'd be a great thing that I'd be happy to invest in.
Right. Right. You mentioned earlier about electric co-ops, and you mentioned they're doing some of the open access model. But can you talk about these non-traditional participants in your business? What are the prospects, and how do they differ? What are they looking for differently than your traditional customers?
It's great. They really don't. Every customer has a way or two that they're a little bit unique. But at the end of the day, it's all about. There's a real need in rural America and look, in all parts of the country, our customers are really excited about providing capabilities that customers want. If you've been in rural America trying to live with a low-bandwidth connection, and somebody will come to you and provide that, I think that's an exciting time. It's interesting for that particular subsegment of the industry because they have a different view. They say, "We already got one set of wires to every house in this county or collection of counties. We'll figure out how to get another one there." So it's just a different mindset.
I think it's also a testament to what great work the OEMs have done with GPON technology because organizations that might have had difficulty with the technology 10, 15, 20 years ago now can tap into a really robust set of equipment and tools that makes them able to operate these networks pretty well. So again, just a testament to how this industry if you'd have told me when we started working on Fios 20 years ago that ultimately all of the innovation that would create in the fiber-to-the-home space would enable electric co-ops to get in the business, I think we all would have said, "That seems pretty far-fetched," right? People forget the initial fiber-to-the-home deployments. The baseline service that was provided was 15 megabits. People were like, "This is really great." And there was always a debate, "Well, do you really need it?
Right. But yeah, I remember the dial-up days even before that. And the use cases, of course, came, especially with the Netflix era in the last 10 years. Wanted to just talk about your financials the last five or six minutes. The one thing that caught my attention is you mentioned that maintenance is a majority of your business. Just help us understand the opportunity on the maintenance side. What's interesting is just passively, fiber to the home requires less maintenance, I would think, fewer moving parts. But curious to hear.
That's right. So what's been interesting and what I and what we think about is recurring, maybe not maintenance, is that part of the business that is just kind of routine, business-as-usual CapEx. A new building gets installed, and somebody's got to take the fiber to that new area, new subdivisions, road moves, cut cables, car pole accidents , just all of the things that go on every day. And people don't recognize how important they are because if we don't do them, people are out of service. So that's an exciting part of our business. And really, the core, as I said earlier, if you think about it, 70 years providing a set of services through probably if we think about that, in 1954, we were probably working on open-wire telephone technology, right?
Then we would have or lead cable, or we would have plastic cable, and then fiber optic to copper, and then fiber optic all the way to the home. The same on the cable side, starting out with networks that didn't have any fiber in them. They were all coax initially. I just think it's an exciting opportunity that there's that recurring book of business. What's interesting is, and the past is no guarantee of the future, but part of what has helped us grow that part of the business is as we've gotten involved with large fiber programs, the capabilities of the fiber going forward has allowed customers to consolidate their supply chains.
We've seen that as an opportunity for market share growth where the scale that we provide customers is more important to them post a fiber build than it was in a previous technology before the fiber build. Just no guarantees, but that's how it's worked over time. We've gotten bigger in that business after fiber programs than we were before because of the opportunity to gain share.
Right. And that sort of recurring costs or recurring revenues might help you on your upfront costs for initial builds. I'm trying to understand maybe when companies enter a new region, there's a level of startup costs. And maybe you can help us with those fixed costs. And then when do you see the revenue sort of come in on your projects?
Yeah. So at our scale now, and that doesn't mean we have ubiquitous geographic coverage, but in most of major metropolitan areas, maybe not so much, New York and Chicago, but pretty much everywhere else, we've got existing facilities. And so we always look forward to the opportunity to increase our revenue density so that if we're there in a maintenance relationship or routine business-as-usual, if we can add a fiber program into that footprint, that allows us to kind of leverage that geography and that infrastructure, not only not just physical, but the knowledge of the permitting relationships, the local DOT, all of the things that you have to be able to navigate well to succeed. So that's probably a little bit different than maybe some of our other peers who are smaller. But we've got pretty good coverage across the country right now.
I guess that brings up the, I think, my first or second topic is what differentiates you is you have that local knowledge too that can go a long way.
Yeah. What I would tell you, it's local knowledge, Greg. And then the other thing is there's so much opportunity out there. If you'd have told me even 5 years ago that you could grow $950 million over a 24-month period, I would have said that might have been a stretch. But what it allows us to do is be really thoughtful about what we sign up for because in all of the opportunity that we're addressing, we want to make sure that we can deliver for customers. This is a business without them. There is no us. Without them, I don't get a paycheck. And so making sure that we figure out where we can best serve customers is really helped by having a broad footprint and a broad set of customers.
That's one of the reasons why this most recent fiscal year, because I just looked at the number, we were below 60% with our top five customers. This is an industry that's been consolidating for a couple of decades, three decades. That was the lowest level since 2001 when we did about $800 million of revenue, now just under $4.2 billion. So we're good.
Right. Diversity.
And more diversified.
Yeah. I guess the last question is then, with all this build, help us with your CapEx profile. You're going to be able to spend more to keep up with this demand. And maybe if I can feed off that, your balance sheet, is there anything in terms of liquidity or access to borrowing capacity that you're going to need to fund this growth?
Yeah. So let's start with the good news on the balance sheet is that we had the foresight, or I might say luck, to sell $500 million of high yield at 4.5% in April of 2021, eight-year note. So really no near-term maturities. We've got a nice credit facility that runs through April of 2026. And so we've got a good cost of capital embedded in the business. And in fact, at year-end, on a net debt to adjusted EBITDA basis, 1.41, so as low as it's been in over 10 years. So we've got lots of financial flexibility. And that's after doing a nice-sized acquisition last year. So we feel good about the financial capabilities of the company. And that's all about supporting customers. When they want us to grow, we've got to be ready with the capital to do that.
Then in terms of CapEx, we love spending money on CapEx. That means we're growing. Typically, we had an opportunity, and we're still trying to work through this. But we were able to stretch our useful lives during the pandemic so we could grow all that because some of the equipment's not available. We've got more equipment coming in now. For the most part, supply chain's pretty well straightened away. So good opportunities to grow CapEx. If CapEx is growing, we're growing.
Great. Well, with that, we're just about out of time. So Steve, thank you very much for joining us. Thanks for everybody for joining.
You too.
Hope you talk soon.
Thank you.