Vertiv Holdings Co (VRT) Earnings Call Transcript & Summary

Good afternoon, and my name is Nigel Coe from Wolfe Research. And thanks for joining us for the last session of the day at the Wolfe Industrials and Transports Conference. I'm joined by -- this is Chief Strategy and Development Officer, Gary Niederpruem; and Chief Financial Officer, David Fallon. Gents, Thanks for joining us this afternoon.

So I think we're going to get just into the Q&A here. And you've been a public company now for over a year. It's been quite a year. You've gone through a pandemic. But one thing that shines through loud and clear is just the underlying strength of the data center markets. And I'm just wondering maybe, Gary, if you can just kick off in terms of what you're seeing right now, real-time and sort of the next 6 to 12 months in your key data center markets.

Absolutely, Nigel. So first off, thank you for inviting us and happy to be here and talk to everybody that's online. Yes. It's one thing to say that you're in a good industry. It's another thing to have demonstrated it. And so I think as unfortunate as that pandemic has been in so many different ways, shapes and forms, for us and from a very parochial standpoint, I think it brought light to a very underappreciated industry. So everybody understood, relatively quickly, the need for data centers and what they do when everybody started to do e-commerce and shop online and telehealth and telemedicine and all that. So I think the pandemic showed a good light on the industry that was generally underappreciated. In addition to that, obviously, we saw really good growth in Q1 of this year to come up to your point on, "Hey, what do we see in real time?" So Q1 would have been 22%, 23% growth in sales and orders over prior year Q1. Backlog is at an all-time high. And when we think about the forecast for the next 6 to 12-ish months, Nigel, we would say , look, the data center space sort of through the cycle, we think, is going to grow 4% or 5%, and we have a stated objective to grow 1.5x that. What we're seeing present today, though, is probably maybe a little bit of COVID bounce back, but we're just seeing a really pretty robust demand environment. Backlog does hit an all-time high. That backlog, all of the backlog that we have today is going to turn in a matter of 12 to 15 months on average. We have really good visibility, I would say, out, probably, 2 quarters of what the year is going to look like. And then from a pipeline standpoint, we can probably see out another 2 to 3 quarters. So we probably have about a 12- to 15-month view of what we think is going to happen. And at this point in time, knock on wood, things seem to be pretty favorable. So the telecommunications companies are still invested in 5G. Enterprise folks are starting to come back online, albeit slowly maturely. And the cloud and colocation folks are certainly continuing to see double-digit growth in almost everything they do. So all said and done, we're blessed and fortunate to be in a good industry. We think we've proven that. And then our ability to execute, certainly last year and as we get into this year, has been pretty robust as well.

Yes. The 4% 5%, is that data center and communications? Or that's just data center?

Yes. So when we look at Vertiv in total, we would say 3% to 4% for the entire portfolio. Of that, the data center probably is on the higher side. So data center is probably in that 4% to 5%. Telecom is probably in the 1% to 3%. And then we have a little bit of exposure to the commercial and industrial segment, which is probably more like GDP. So when you blend that all together, we say Vertiv in total, through the cycle, is probably 3% to 4% with data centers probably closer to that 4% to 5% range.

Yes. Okay. Just that a lot of your competitors highlight data center as more like a high single-digit opportunity for the next 3 to 5 years. And certainly feels like there are a lot of structural drivers here that maybe suggest that 4% to 5% might be on the low side. Any thoughts on that?

Yes. I hope I'm wrong and when I'm wrong, I hope I'm wrong that it's higher than what we're prognosticating. I think for this year, we will see that 4% to 5% creep up. I think it will be -- I don't know if it's going to be 5%, 6% or 7%, but I think it's going to be a little north of 4% to 5%. but if you'd use last year as a baseline and then go out 3 years or so, that CAGR probably is still going to be 4% to 5%. So I think mid-single digits through the cycles about, right, this year may be a little bit more on the elevated side.

And [ this is ] not a bad place to be, to be honest here. Your outlook -- and just come back to the 4% to 5%, since -- maybe a little bit more. If we do have a, I don't know, 6% to 8%, 8% to 10%, whatever your competitors are saying, would that require any additional sort of investment spending be a CapEx capacity investment spending? Or would that volume upside accrete to the bottom line? I mean I'm just wondering kind of how levered your investment spending is to growth.

Yes. I can start and then David, certainly, can opine on this as well. So we've been pretty resolute, and David has led the charge when we talk about fixed cost constant. So even in a growing top line environment, all of the G&A pieces of it from a fixed cost standpoint should be relatively flat. Obviously, we've been very well-known that we're investing in engineering and sales and marketing. I think this year, that number is probably $65 million of investment or so with 2/3 of that roughly in the engineering world and 1/3 in the sales and marketing world. So we're going to continue to invest in that until we can get to engineering up to around 6%, which will probably take another year or 2 to get there. In order to deliver the plan this year, I don't think that we need to add any structural fixed cost. But I think what you'll see us do is continue to, one, make sure that we're investing in the right tools and technology and IT, which sometimes it's an FX cost, sometimes it's a CapEx cost. And then from a facility standpoint, we're always just going to want to make sure that we're positioned in the right regions for the regional growth. So we have enough between what the Vertiv operating system is doing in the lean initiatives. We have enough square footage. We have enough fixed cost to be able to deliver the plan this year. I think from our standpoint, it's really more about what do we want to invest in today that will affect positive growth in year 2, 3 and 4 from now. But David, please, I'll have you comment on that question as well, if you don't mind.

No, I agree with everything you said. And specifically, from a manufacturing capacity perspective, we believe we certainly have sufficient capacity this year to fill what we need to deliver to customers. If there would be a ramp-up in demand, as you mentioned, Nigel, above the 4% to 6% or what have you, we believe there is enough inefficient use of floor space in our facilities, that we would be able to create that capacity from a machine utilization perspective or floor space utilization to create that additional capacity without significant CapEx. So we wouldn't need significant equipment or facility expansion CapEx to meet that higher level of sales.

Great. And this -- I was going to touch on this a little bit later on the conversation, but now might be a good time. Factory productivity is one of the key levers of cost reduction or better margins going forward. And so I'm curious, you just raised the point about inefficient use of space. Where are we? And what is the path towards better factory utilization levels, better factory efficiency levels? Maybe just talk about some of the key buckets of cost reduction that you see over the next 3, 4 years and where we are in terms of progress amongst those -- along those paths.

Yes, absolutely. And any time we talk about the opportunities going forward. One of the initiatives that effectively touches all of them is the Vertiv Operating System, VOS. And it's patented off of the Honeywell operating system, which we shamelessly borrowed from Honeywell. We actually spent about a week on Honeywell's campus last year, pre-pandemic, to learn the Honeywell operating system. And I would say we're still in the early innings as it relates to the implementation instead of a big-bang approach, taking what we've learned and trying to push it into all 20 facilities. We've selected 2. And we've made great progress in those 2 facilities in 2020, and we'll be expanding that to some additional facilities this year. But I think a baseball fan asked what inning are we in as it relates to Boston. It's probably still the top of the second inning. So it's still really early. But the positive results we've seen have been very tangible. And it touches all the elements of lean that you see in other more mature manufacturing organizations. We've seen a freeing up of floor space, and a lot of that is related to inventory management. So our work in progress, finished goods have been reduced significantly, and our lead times have tightened as well. So it's still early on, but we see loss not only impacting what's happening on the plant floor. But also what's happening in the back office. So one of the initiatives that is very key to our margin expansion is this philosophy of fixed cost constant. And I could use the rest of the time here today to talk about it, but I won't. But it's as simple as it sounds, right? It's -- we have about $1.4 billion of fixed costs. Philosophically, our goal is to keep that flat going forward. And we will be using, and we have been using the tools of VOS to introduce productivity savings. So we're going to have increases in fixed costs, whether it's related to merit increases or other forms of inflation, our goal is to offset those higher fixed costs with productivity. So just that concept of fixed costs, which we were able to successfully do in 2020, and we're well on our way doing the same thing in '21 with the exception of maybe some higher R&D and sales costs that Gary mentioned, that's going to be a huge lever to take our operating margin, which is about 12% today to 16% in the intermediate term, and that's where we roughly think our competitors' at. And in the long run, there's no reason why we shouldn't be able to get to 20% plus.

And just to be clear, that 16% on an EBITA basis?

Yes. Effectively, yes. What we're defining as adjusted operating profit, which removes the impact of the amortization of the intangible stuff.

So on an EBITDA basis, we're in the -- sort of the 17% handle, 17.5%, 18% type of range?

Correct.

Yes.

Okay. All right. So 20% EBITDA would be definitely a very decent margin. And by the way, if you want to talk about fixed cost neutrality for the whole conversation, I'm okay with that. Maybe others wouldn't, but I'm plenty okay with that. Just thinking about -- so just to be clear, R&D is fixed cost. You're classifying now the fixed cost. That's within your bucket. So obviously, this year, we're not covering that step-up in R&D. But going forward, you're confident that the move from, say, 5% -- roughly 5% R&D today, 6% R&D, that will be covered by productivity.

Yes. So I would say the lone exception that there will be, as it relates to fixed cost constant will be reinvestment back into the business. We're not going to necessarily create artificial financial targets that would necessarily impact the longer-term growth of the company. So R&D is one of those exceptions, and we've been pretty transparent that there's about a $65 million headwind this year that we are not covering with productivity and probably $35 million of that is R&D, $30 million in sales and marketing to support the products that are being developed. And there will be a further ramp-up of that R&D. I think the total R&D is probably $265 million for '21. That's about 5.4% of sales. Our goal is to get to at 6% in a steady state business. So all other things being equal, if you take our top line guidance for '21, that's $4.9 billion. It's another $30 million of additional R&D cost to get to that 6%. Now of course, it's the percentage that remains flat thereafter. So the R&D spend will continue to ramp up. Once we hit that steady state, our goal will be to offset any increase in R&D, any increase in merits or other inflation with productivity.

Great. And then, Gary, moving to you and staying with the same topic. The ramp-up in R&D spending. You provided some great examples of what's kind of the new product vitality you're seeing today. Where do you think Vertiv needs to really improve competitiveness across the product portfolio, be it geographic, be it across the product portfolio, where do you think the investment dollars? And where do you think that's most needed?

Yes. I think it's really balanced right now, Nigel. So I would say less of -- I'll frame the question just a little bit different. I'll tell you in sort of where it's needed, just early where we're pointing it, which is the most important. So we have a tremendous amount of those incremental dollars flowing into power and thermal for the cloud and colocation customers. And what we continue to find time and time again is this is an industry, and those are classic customers. They want to be collaborated with. So you're not going to go to them and say, "Hey, I have a white pencil, I'm going to sell you. This is what you're going to take. It is very much, hey, what problems are you trying to solve? We're thinking about it this way. They say, no, I'd like to do it this way. These are my constraints and you go back and forth and it's that collaboration. That collaboration takes time and people in deep domain knowledge, which is what we have. So there's a lot of dollars flowing into power and thermal for the largest customers to be very, very collaborative with them. That's number one. number two is there's a lot of dollars still flowing into making sure we have absolutely the right product portfolio for the IT channel and edge. So small systems, small integrated solutions. There's a lot of individual point products that we think are going to play very, very well with small UPSs or small power distribution units. So we're making sure we have a world-class lineup of products for edge in the IT channel. And then to David's point earlier, we're wrapping marketing and sales around all of that. So the last thing we want to do is I have the world's best product and solution lineup and not being able to tell anybody about it. So we're making sure we have the right application engineering, technical sales force, marketing, demand generation, sales enablement, all of those things that make a successful launch, successful. That's what we're doing. So it's really those 3 legs. I mean, there's lots of other things, obviously, but they're trying to distill down into 3 big buckets. That's probably the easiest way to digest that.

Okay. That's great. That's great, Gary. So it feels like a big focus is being able to customize solutions for your customers. Is that fair?

Yes. I think that's very fair. And it's interesting because we still go-to-market with platforms in building blocks with all of our products. But what everybody wants, is that last 40%, 50%, 60% or so. They want that tailored or customized. So not everything is 100% bespoke. I don't want to give people that impression. But there still is this philosophy of you can have the global building blocks up to a certain level, but then once you get to that level, you do need to have some degree of customization, particularly for the cloud and colocation customers. There's no doubt about it. And it is -- the company historically got burned because it was very much, "We're going to design something in the back room. We were going to tell the industry, this is what they need to buy." And so we learned that lesson a while ago, and we're going to -- we work like heck every single day to make sure we don't repeat the same mills from a decade ago.

Right. Right. So voice of the customer is very important. So a question we get, a recurring question. And it's a question we get for other companies as well. But what is the key kind of competitive barrier to stopping new entrants coming into your markets. What is your key competitive advantage? How do you differentiate versus your competitors, but also new entrants? And we've got our series, our responses, but I'd be curious if, maybe, Gary, you could like kind of give your thoughts there.

Yes. So I think there's 3 or 4 things we'll hang our hat on every day on that one. So one is, I think there's a little bit of, "proof is in the pudding," almost that. There aren't too many new entrants you see come into thermal and power and the service and the rack PDU world, that just -- it does not happen all that often. So one is, I think we've proven to demonstrate that because of that, there must be barriers to entry. Now why is that? I think number one is you have to have deep domain knowledge to operate in this space. You can't walk into large customers that are going to spend millions of dollars on your product. And let's remember, what we do is we provide power and cooling in all of the environmentals to keep the vital applications up and running. So one of our systems has an issue and that takes down our customers' data center, that has consequences for them. So they're not just going to trust it to the low-cost bidder. So there's a very real reason why we call vital applications, and that's what we support. So because of that, you need to have deep domain knowledge. We couldn't walk into a customer with a band of merry men and women who just -- we picked off the street and said, "Hey, trust us, we're good for it." It doesn't work that way. You have to have deep domain now, which is number one. Number two is, you really do, at this day and age, you need to be not only global, but what we call is -- we call it local everywhere. And so customers are used to just deploy in the U.S., today are now deploying in Latin America, or Southeast Asia, or parts of China. And you can't do that unless you are not only global, but you are local everywhere. So that's number two. With that comes in region, 4 regions. So all of the salespeople, all the application engineers, all of the manufacturing footprint we have is in a region, 4 regions. So that helps tremendously. Number three, coupled with 1 and 2, is you need to have really good products that you're producing that are highly efficient that are -- that do have a small footprint that can be deployed easily. So there is a lot of technical know-how within the product. And then lastly is the service organization. So $1.4 billion of our $4.9 billion is service. We have 3,200 field technicians all around the world. Things aren't always going to go wrong. It's how you respond to that. And just take a look at the pandemic. If we were -- said we were global, but all of our service technicians were in the U.S. and a customer had a problem in China, well, how are we going to solve that? You can't do that unless you are local. So having that service footprint is unbelievably critical, and it's still true to every class of customers. Now different class of customers may want different service levels, and that's okay. But I think it's between the deep domain knowledge, being local everywhere, having technologically advanced products and having a service organization, those are the 4 things we keep coming back to time and time again.

And is the service organization, I mean, first of all, Gary, can you maybe kind of size your service footprint compared to some of your competitors? I believe that you do have the most extensive service footprint amongst your key competitors. And then is service more impactful for certain customer types? I know it's a bigger business in the U.S., roughly but is it more impactful in kind of the on-prem versus off prem? Any color that would be helpful.

Yes. So I'll take your first part first. So you're right. We -- for what we do for our basket of goods and services, we do have the largest service organization focused on digital critical infrastructure. Some other companies have service organizations that physically may be larger, but they're playing in lots of different spaces. So that's not really a fair comparison. If you just baseline what we do in the service organization we have, we are, bar none, the largest service organization in our industry. So that's number one. Number two is the service business is really pretty proportional in every region of the world. So it's not like 1 region of the world, we do $1 billion; in another region, we would do $100 million. Some are bigger than others, certainly, but from a percentage of overall revenue. The service business is pretty close in each one of the regions, which just is another example of how important it is. It's not like one region values up more than another. From a customer set standpoint, there's no doubt that the service business has more enterprise-type revenue attached to it. But a big fallacy is people think that the cloud and colocation customers just don't need our service. And that is just flat out incorrect. They might want a different level of service. They might say, I only want you to visit me twice a year versus 4 times a year. There may be more of a partnership and collaboration there. But absolutely, we have service contracts with almost every one of our customers in some way, shape or form. So it -- that almost $1.5 billion business with those 3,000 service technicians are huge. Not only from servicing the customer, but the other thing we get, back to our earlier dialogue, was we get voice of customer from them. So they provide real time feedback. Customer had this issue. I had a hard time installing this. We could make this easier for the customer if we did this. So it's a real-time feedback loop that we get from that service organization as well.

Thanks Gary. I think we'll go to the questions from the audience here. I've got 1 question in the box here. So folks, if you go any more questions, please log those now. The question is, how do you ensure that the Vertiv Operating System really takes hold? Is a week at Honeywell enough to learn the system? what has to be done to ensure VCS -- VOS is successful?

Yes. I'll take that one, and Gary, certainly, fill in the blanks. But I would say it's -- when you look at an operating system, whether it's implemented at Vertiv or Honeywell, it definitely requires a concurrent cultural shift. So the one thing that is very clear with the implementation of VOS is that it is not a pure manufacturing initiative. Even though there's a lot of focus on launching it at a couple of plants, I can tell you that the Vertiv Operating System has been launched globally. So very specifically, speaking, somewhat parochially, from a financial function perspective, we just had a virtual 5S day last Thursday. So 5S, which is a tenet of lean, which is -- we think about it as organizing, having tools that you need at your workstation to do your work and having them in a place where you need them to be. We held a global virtual 5S day for the entire finance organization last Thursday. And so we had 600, 700 folks virtually clean up their desktops, clean up their email. And there's some practical significance there, but it was more symbolic in the optics of it of being part of not only an operating system that impacts the plant floor, but the entire Vertiv family. So and to answer your question, is 1 week in Honeywell sufficient? It clearly isn't. We have -- yes, we kind of looked at that week as kind of the kickoff for what we needed to do. We have hired significant resources that are very specifically focused on the implementation of the Vertiv Operating System, both within the factories and the back office. And so we have a robust team, many who have past Honeywell experience. But this is a concept that we launch every internal staff meeting with some discussion about safety, which is very critical for the Vertiv Operating System. And it's something that we talk and live on a weekly basis now. Oftentimes, a lean is -- or a lean journey is something that effectively you never quite get there, and it could be 10, 20 years, you're still continuously improving. We think the continuous improvement will be ongoing, but we do have an internal goal by the end of 2022 to be in a spot where we're seeing very tangible improvements across the board, across all functions. So we probably have another 18 months. And even though this is something that can't happen overnight, we do have some internal benchmarks as far as when we think we'll start to see some critical benefits from it.

Okay. Good. But in terms of making sure there's no backsliding at the factory level? I mean how do you ensure that?

With metrics. And we have very robust visual metrics, which is a key part of lean that are reported on, on a continuous basis. This isn't something that we look at on a quarterly basis or semiannual, even on a monthly basis, this is something that we track metrics on a daily basis and very visually. And that process is probably more robust at the plant level right now, and there's very heavy communication and reporting all the way up to Rob Johnson's level. But we're starting that metric and that reporting process through all the functions right now and should be further developed on that front by the time we exit 2021 and fully there at some point in '22.

Thanks, David. I've got a few more questions here coming in via text. Are there any large or emerging competitors that are trying to compete on price? And I think this comes back to -- to your point, Gary, on barriers to entry. What's your protection against that?

So I would say, for the most part, if you look at the pie charts and how we break down the distribution of our product lines and the competitive nature, roughly probably, to us, and if you include the 2 or 3 other multinationals that are in the space, we probably got 40% to 50% market share on average in any given product category. So obviously, that set is very, very rational, very, very tried and trued. No issues there. The predominance of the other 40%, 50%, 60% in any product category are really local or niche or point product competitors. None of them are really new for the most part. It's just that they say, look, I'm a thermal company and I operate in Germany only. Okay, great, but they're not going to do anything more than that. I'm a power company that's working in India. Okay, I'm not going to do any more than that. I'm not going to export there. So we don't really see a lot of it. The only one that sometimes becomes the elephant in the room is Huawei. And what we've seen from them is they are probably I don't know if they're more rational today than any given point in time. But there is a lot of -- the disruption in that Huawei cause primarily hit the industry 6, 7 years ago. And ever since that time, they've gotten a little bit smarter. We've learned how to compete. They've actually made us a better competitor in a lot of ways. So we really don't see, Nigel, anybody coming out of the left field that says, "I'm just going to -- if your price is $100, my price is $50. Now deal with it." That's -- it doesn't act that way, and we don't really see that in any one of the product lines in any one of the geographies.

Great. There's one more question from the audience. But this will be a good time to take a pulse on pricing. And yes, I think there's a perception, probably going back to the Emerson ownership days, that pricing in this market is intrinsically weak. And in the [ phase and ] time, that creates some risks around margins. So maybe just address that, Gary. How do you see pricing power? Maybe just dissect that between enterprise and hyperscale or colo and what is your confidence that over time, you can cover the inflation that you're seeing?

Yes. No, we've been dealing with that, obviously, for real time. So I think a couple of things. So David says it very well that prior to 2017-ish or so, there really wasn't a focus on pricing, and I think the perception was exactly what you said. And maybe that was part perception, maybe it was even part reality back then. But we kicked off some pretty serious pricing initiatives several years ago. And if you look at -- in 2019, we got somewhere between $20 million to $25 million of positive price. In 2020, it was probably $20 million of price. And it's from a number of different tactics and initiatives, and some of it is easy as well, you raise list prices every year in some places. But that don't only get to so far. After that, you need to become much more judicious with your multipliers and your discount levels and how are you handling rebates? And let's make sure that we have scattered thoughts of, "like jobs to like customers," in the same region in the same category. We weren't doing those types of things before. So those are not out things that we do almost every day, maybe not in every region, but sort of Americas is the most mature. Europe is the second mature. Asia is the least mature, but they're all significantly advanced from where we were a couple of years ago. So we have 2 years of demonstrated track record of getting positive price. When you look at this year, in our Q1 report, we had a Q1 presentation. We had there was about $25 million of headwinds and $10 million is going to be offset by price. So I would say 2 things. One is that $10 million clearly is the floor. I mean there is more upside and more runway on that number even -- and we sort of even knew that at the time, but we wanted to be a little bit conservative on that point. So there's clearly more upside and runway in-year than that $10 million. And David and I review pricing and the commodity stuff probably got -- it seems like it's 8 days a week at this point in time. That's because we want to make sure that we're doing all the right things. And I am sitting here right now, I am really, really confident that we will be able to recover the vast, vast, vast majority of any inflationary stuff. Now some of it -- you heard there's a lead lag there. So some of it may hit the P&L or maybe a quarter offset, just based on the time that you see the cost versus when we can get the price. So there might be a quarter lag there. But in terms of picking it up over the course of the year and then looking into next year, some of the commodities abate a little bit. Most times, our pricing is pretty sticky. So I think that even if we miss it by a quarter or so offset this year, that will turn into a nice tailwind for us as we go into next year. So I feel -- bottom line is still really good, about being able to get the vast majority of it, maybe a little bit of a timing dimension associated with that, but I think overall, we're going to be in pretty good shape.

Just to be clear, so when you say the vast, vast majority, do you mean actually in this year, so that $25 million of incremental inflation do you think you can capture that this year? Or are we talking here about catching up on a run rate basis by the year-end?

Yes, I don't want to get in too far into probably anything different than what we guided to already. I would just say, whatever the inflation number is going to be by the end of the year, maybe we don't get all of that in here, but we will certainly have recovered it on an annualized basis when you look at it over a 12-, 15-month window.

Got it. $10 million of price, $25 million of inflation. I understand it. And then a question here, how do you think about the right balance of centralization versus decentralization compared to Emerson's ownership, which, for the benefit of those on the audience, was very decentralized.

Of course, very -- wait, did you say decentralized or centralized?

Yes.

Well, you broke out. One more, say -- I just want to make sure I heard the question.

The question is what is the right balance of centralization versus decentralization in businesses? I think the -- kind of the view that Amazon ran a very decentralized business model.

Okay. Yes. So I would say, organizational models take a little while to find their footing. And 4 or 5 years into this Vertiv venture here, I think we feel really good about the balance we have struck. So we have salespeople and service people that are really, really local with the customers every day. And that's where those decisions need to happen. David and I might get an escalation because we're going to quote a big job in Australia. But at the end of the day, the people that know that price point are the people down in the Australia. So we want to empower the local people to make those right pricing decisions. From a product management and product road map standpoint, though, there is enough overlap in the regions that we want to make sure we get those right building blocks established, so there's governance around the building blocks at a -- on a global level. And then that customization occurs sort of based on a customer. So there's no one size fits all. It's not either fully decentralized or fully centralized. But I think more and more, we gravitate towards we want to do business how our customers want to do business. So if you're a large multinational cloud and colocation customer, then we're going to service you locally. But we know at the end of the day, a lot of the decisions are made in headquarters, so that's how we'll move. If you were an enterprise customer and you're operating in Southwest Texas, okay, but we want to make sure that people down in Southwest Texas have the right ability to offer at the right price point. But when it comes to the product lines, we'll have a little bit more of a global bent on that. When it comes to where we manufacture, the supply chain, that would be a little bit more global decision-making as well, just to make sure we don't get too over-indexed in any one given region.

But then, I mean, I think the challenge always is to act like the centralized business model, very entrepreneurial business model on the front end, but then more centralized on the back end in terms of functional costs and sourcing and things like that. So do you think you got that balance right at this point?

I do. And I wouldn't have said that certainly not 3 years ago, maybe not 2 years ago because it does. It takes a while for the organization to understand the checks and balances there. We hold monthly reviews with each one of the businesses to make sure there's nothing that's going on that skews. There's very real-time updates there. That's in addition, all the functional reviews that David or I may sit in. And then the other thing that we do that is -- really help catalyze all of that as we all have the product people and the regional people do what we call an interlock. And they will do that interlock on hey, what is this customer saying? What's the voice of the customer? How does that inform in the road map? When is this going to come to market? When should I start pushing that? So doing those types of things is invaluable, but it definitely takes a year or 2 for the rhythm of the business to kick in, and we've been at it for 4 years now. So I do feel really good that we have that right balance.

Great. You've been very generous with time. I've got 2 more questions, and -- but I think we'll wrap it up. You talked about edge, Gary, in the discussion. That's a huge topic, obviously, of interest for many investors, especially just given how flat to kind of challenging COVID must have been and -- enterprise must have been -- since it's so long now, there are some out there -- that have a view that enterprise can actually grow much more comparably with colo and hyperscale going forward. And there are some that say, look, the acceleration from on-prem to off-prem is going to accelerate from here. Where do you stand in that view? And maybe somewhere in the middle, I don't know, but do you think edge could be that impact for take growth rates up into the high-single digits?

Yes. It's funny. It's almost like a religious debate, and it's fun to be able to listen and talk to people on all sides. So my perspective, I think this is what the strategy of the company has been around, specifically in enterprises. We don't have our head in the sand from the standpoint of -- the traditional enterprise-type customer is going to continue to move applications and workloads more to the cloud and co-location companies. I mean that's just a fact. So our head is not in the sand from that standpoint. But what they do have is there's still a hot -- the data is growing so fast for these folks that they still need to keep their own data centers open. They're still refreshing them. There's still a lot of FINRA and HIPPA laws and said and proprietary information, they have to keep on-prem. So for all of those reasons, the core traditional large data center and enterprise, we think is probably going to shrink low to mid-single digits. So the classic data center, enterprise customer probably shrinks low to mid-single digits. That is probably even conservative. There might be a little bit better than that, but that's what's in our planning thesis. But the offset to that, Nigel, is exactly what you just said is the advent of the edge. A lot of the edge applications manifests itself in a number of different ways, but one of the biggest predominant ones is in the enterprise customers. So all those distribution facilities, all those point of presence, all those IT costs, all the manufacturing facilities, all of that is being upgraded and will continue to be upgraded. So we think the edge part of enterprise is going to grow at 15%, 20%. Now it's albeit it's a smaller piece of the enterprise business. But when you merge that altogether, enterprise, we think, from a market standpoint, is probably flattish. And our growth is going to grow low single digits in that category, knowing those dynamics. So at the answer is definitely the hybrid world is very real. You'll have on-prem, you'll have off prem. People will continue to move things to off-prem, but there's just so much data being created that there is always going to be a need for enterprise data centers for a lot of those customers. So that's how we sort of bifurcate enterprise. When you aggregate it together, we think it's probably a flattish environment, and we certainly are pretty bullish about taking a couple of points of share at the same time.

Okay. So somewhere in the middle. And then, David, on free cash conversion was sub-100% this year. Maybe just run through some of the reasons why that is. I know there's some tax payments kind of to consider there. But longer term, is there any reason why Vertiv shouldn't be 100% free cash conversion?

Yes. And that is our goal. When we look at our multiple, we think we probably are trading at a discount anywhere from 15% to 25% compared to peers, whether direct or indirect. We think it's driven by 2 things. Number one, is the margin gap that we talked about earlier. The other is that free cash flow conversion and our absolute goal is to get that to 100%. There's no reason we shouldn't get there. Right now, CapEx is a little bit elevated. I think in that $95 million to $100 million range is -- that's outpacing our depreciation by a good $15 million, $20 million. So that's part of it. And we have additional opportunities on the working capital side. So we're aware we're not to where we want to be, but I can tell you management as shareholders, but we understand the importance and getting to that 100% free cash flow conversion.

Great. Thanks, David. I said that would be the last question, but there's one more coming in from the audience, which I do want to get to. And the reason being is that normally, in these fireside chats, the audience goes to sleep and there's no questions. So the fact that we've had 5 questions from the audience is actually very good, so I want to make sure we address this. As a quick one. Any concerns on U.S.-China tensions?

By being a wake, that wasn't my mom that sent it into your -- I hope it wasn't. it?

I'll check later on.

Yes. Our business in China is really pretty robust. It grew substantially last year. We are very cognizant of U.S.-China relations. But for those of you who don't know, the history of our business in China was we bought the Huawei Power business, Emerson did back in 2001 or 2002, I think. And so to this day, we act very local. We have also -- we have our supply chain there, engineers there, sales force there, manufacturing there. And if anything, we are continuing to make sure that we act more and more local as time goes on. So that has not caused any disruptions. We constantly evaluate it. We had an MBR with them just last night. That was a topic, but so -- we're keeping our pulse on it very, very closely. But we're in this nice segment where we're not IT security space. We're not providing bits and bites, but yet we are strategic enough that they want to make sure that they pick a trusted partner. So it's this nice sort of mix that we're fitting in. So watch it very closely, Nigel, but we haven't seen any signs of disruption because of any geopolitical issues at all.

Sounds great. Thanks, Gary. Thanks, David. You've been very generous with your time. Appreciate the conversation. Thanks for being here, and look forward to catch up soon.

Thanks.

Thanks, Nigel.

Thanks. Nigel.

Thank you very much. Bye-bye.