Data Center Construction and the Shift Towards Modular Builds | Work Done Right with Luke Kipfer

Luke Kipfer joins the Work Done Right podcast this week to delve into the dynamic world of data center construction. One prominent theme highlighted was the industry’s significant shift towards modular construction methods. Recognized for their speed, cost-efficiency, quality control, and safety enhancements compared to traditional stick-built approaches, modular designs are gaining traction, largely fueled by standardization, enabling the efficient deployment of infrastructure across diverse locations.

Luke also discusses the critical role of Northern Virginia as a central hub for data centers, with nearly 70% of global internet traffic passing through this region, a fact that becomes visually evident when flying into Dulles International Airport. Moreover, Luke emphasized how technology is reshaping construction safety, citing innovations like ergonomic tracking and safety monitoring tools that not only mitigate risks and safeguard workers but also foster a cultural shift towards prioritizing safety within the industry.

 

About Luke Kipfer

Our guest today is Luke Kipfer. Luke is the vice president of data center development and construction for American Real Estate Partners and Powerhouse (AREP Powerhouse) Data Centers, where he is responsible for planning and execution of all data center projects.  

  

Luke has more than 15 years of mission critical experience leading multimillion dollar data center design and construction. Before joining AREP Powerhouse, Luke served as Regional Director at Directline Global, where he oversaw project management and operations for several of the world’s largest hyperscale data center sites.  

  

Prior to Directline Global, he was the director of construction at Markley Group, where he managed all aspects of design and construction for New England’s largest co location, mission critical telecommunications and data center facility.  

  

Luke is a certified project management professional and holds a B.S. in operations and project management, as well as an MBA in project management.  

Top 3 Episode Takeaways

  1. Shift Towards Modular Construction: There is a growing trend in the data center industry towards modular construction methods. Modular construction offers benefits like speed, cost-efficiency, quality control, and safety improvements compared to traditional stick-built approaches. Plus, standardization helps meet the rising demand in the data center industry by streamlining processes and allowing for efficient deployment of infrastructure across different locations.  

  2. Virginia As A Central Data Center Hub: Luke references a metric that approximately 70% of the world’s internet traffic flows through two different counties in Northern Virginia. This showcases the growth of data center construction in that region, which can be witnessed from the sky when flying into Dulles International Airport.
     
  3. Technology’s Impact on Construction Safety: Technology is playing a pivotal role in improving safety on construction sites, particularly in the data center sector. Innovations such as ergonomic tracking and safety monitoring tools are helping mitigate risks, protect workers, and drive a cultural shift towards prioritizing safety in the industry.

Episode Transcript

Wes Edmiston: 

Luke, welcome to the show.  

 

Luke Kipfer:  

Thanks, Wes. It’s great to be here.  

 

Wes Edmiston:  

Yeah. Good to have you. The data center world has been booming here recently, especially since there’s a lot of computing needed with the advent of AI and how quickly that has boomed up.  

  

But even before that, data center construction has been really over the last 10-15 years, as you well know, been really on the rise. So I’m just kind of curious, how did you end up kind of falling into the world of data center construction?  

  

And really when did you get into construction in the first place?  

 

Luke Kipfer:   

Yeah. So I started construction, really right out of high school. So graduated high school. I had family that was in the trades. My dad worked for a mechanical contractor.  

  

I would say I didn’t have a clear path or direction as a 17 year old graduating high school. So my dad said, well, you’re not going to sit around the house. You’re going to come to work, right?  

  

So there I was, straight out of high school, with hard hat boots on the site, really, which honestly turned out to be the best thing for me, because, honestly, going to school at that time in my life, that education probably would have been wasted.  

  

So been in construction literally my entire professional career. Specifically, we did a little as mechanical contracting. We did a little data center, mission critical work in terms of we were the libert vendor for the area, so computer room, air conditioners, some specialty AC systems.  

  

I went specifically to 100% data centers in 2012. So it’s been a little over ten years that I’ve been 100% dedicated to data centers. And now that I’m in the industry and then I’ve been doing this, I wouldn’t want to build anything else right now.  

  

Wes Edmiston: 

Yeah, the construction industry, oftentimes whenever we get into it, it’s maybe begrudgingly. We have this idea of maybe it’s a temporary thing, but it is an infectious sort of thing, and it is addictive once you get into construction, especially kind of like you’re saying about the mission critical work that you do now.  

  

Luke Kipfer: 

Yeah, it’s one of those professions that’s rewarding. Right. I get this sense of gratification when you see a project completed, when you can drive by a building and say, hey, I built that. Right. So I think it’s a great industry because there’s tangible benefits and rewards to what you see and what you’re working on every day.  

  

Wes Edmiston: 

Yeah, I agree with that completely. I think the number of hours that we end up working whenever we’re on site, it kind of forces you to be all in to whatever it is that you’re doing. So that kind of monotrack focus that you end up taking whenever you’re in the thick of it, I think it kind of helps to push everything out, and it almost like forces you to fall in love with the work.  

  

Yeah. You fall in love whether you like it or not, right? 100%. So just talking a little about what it is that you guys do now at ARP powerhouse and what you specifically are doing, you all have a very different way of approaching these projects.  

  

I think it’s a brilliant way of approaching them, but it’s very different than the typical, hey, a client comes and approaches you. A customer approaches you. Here’s the project we want you to build.  

  

You end up building out all of the specs for it hand in hand with the requirements from the owner and kind of deliver it to turnkey maybe from what it is that they specified on the front end. So can you describe a little about kind of how it is that AREP powerhouse approaches these projects and data center construction?  

  

Luke Kipfer:  

At powerhouse, we’re really a real estate solutions provider at our core. Arep has been a developer for over 20 years. So we’ve spun off Powerhouse specifically as our data center platform. But as you said, the growth of the data center industry, it was just growing by leaps and bounds before some of the new technology.  

  

And AI is just one of the many new technologies. AI is the latest great buzzword, and it is really driving a lot of change in the industry. But we’ve built this platform to be able to deliver its scale and its speed.  

  

So ten years ago, a 1 MW build could be a big data center build. Five years ago, a ten megawatt build was a good size build. We’re looking at single buildings that are well over 100 MW plus. And a lot of these, what we call hyperscales, these major users, one building is not enough.  

  

They’re looking for multiple building campus, 100 plus acres, 100 MW immediately, ability to scale up to there’s gigawatt campuses, under planning. So when you look at the scale, a lot of organizations just aren’t built to be able to deliver the real estate side of that.  

  

So a lot of these hyperscale users are great at once. The structure is built to be able to go in and fit out their Ups systems and their generators and the specialty cooling systems. But there’s a lot of front end work in terms of finding the right sites, getting the utilities on site, making sure it’s zoned properly, being able to do that early site development to put these buildings up.  

  

So that’s what we do. We go out, we handle the site selection, we handle all the utility coordination, so we’re able to literally shave years off of projects. Once one of these hyperscalers has a need that says, oh no, we have this specific workload and now we have to find a home for it, they don’t have time to go out and find a piece of dirt and go through that whole process to be able to get it lined up for them.  

  

They just want to start racking and stacking servers. So we go out, we find the land, we’re able to take down the land, we get it zoned for data centers, we get the power there, we get water there, we get fiber there, we do all the site development, we build the building.  

  

So when a lot of these hyperscalers are looking to deploy again, we literally shave years off the time that it takes for them to go from a need within their organization to be able to stand up a data hall.  

  

Wes Edmiston: 

Yeah, I mean, especially like you’re saying, when they have need, they have instant need. So you guys are able to really just kind of have a lot of that front end work done to where you’re able to buy the one thing that money really can’t mind, which is time.  

  

Luke Kipfer:  

Yeah, absolutely. So we’re able to work with the supply chain and we do a lot of these powered shells. And ultimately in the right situations, we also go out and build the infrastructure as well, too.  

  

So going out and preordering chillers generators, cooling equipment, all these specific long lead items that keeps these people from being able to deploy. And frankly, the need for this new space is almost insatiable right now as the data requirements grow.  

  

Wes Edmiston: 

Yeah. So you guys are coming in, you’re getting a piece of land ready, bringing in everything that needs to be brought in, and potentially standing up the building and getting some additional aspect specs of it ready so that when somebody comes in saying, hey, I need to construct a data center in this approximate location.  

  

You guys are already leaps and bounds ahead of where they would be otherwise. But something that comes up on every single project, seemingly depending on who your different clients are, is every client has some kind of minute differences in their specifications.  

  

We’ll say even the material that they’re using to bring in some of their utilities, whether it’s water or what have it is or the specifications for that piping, the specifications for your building. So how is it, then, that you all are handling this almost one size fits all sort of solution?  

  

How are you having those conversations with the potential owners is the question that I have, because I personally see value in standardization, blanket standardization across the world to cut down on cost and schedule and even quality differences later on.  

  

But not everybody’s so easy to work with. So how do you guys kind of manage those conversations? And also even how are some of these recent changes, like, focus toward AI. Changing how it is that you approach these buildings?  

  

Luke Kipfer:  

Yeah, I honestly love to take credit for it, but really, we have amazing design partners in the industry, and one thing that we’ve done as powerhouse is standing up this new delivery platform is that our partners, whether it be on the design side, whether it be on the GC side, the.  

  

You know, we’ve specifically gone out and use the same teams that these hyperscale providers use directly. Right? So there’s really three to five top architectural engineering firms that build these kind of campuses.  

  

There’s three to six or seven or eight GCs, depending on what they’re a little more regionally based. But we’ve gone and partnered with the people that are building Direct for these users and we’re able to take a lot of those understandings, those lessons learned out of that so that when we go to design a building, we’re able to check through and say, hey, these are our six clients.  

  

These are the guys who are taking down the most data center space in the US. Does this design work for client X, Y and Z? Right. Does the floor loading work? Is our roof wind rating right? Do we have room in the mechanical yard for the equipment and the generators that they’re planning in?  

  

So sometimes it leads to a little bit of overbuilding because we need to build to a spec that any one of these users is going to find acceptable. But I’ve got to say, it’s really our partners within the industry that we rely on to understand how these different users deploy their compute and then ensuring that our design aligns with that ultimate end goal.  

  

Yeah. And realistically, is it similar to the way the automotive industry, for instance, it’s more expensive now to build out vehicles with crank windows than it is power windows because every vehicle has power windows in them these days.  

  

You know, what I’m saying is it almost more cost efficient and effective in order to build everything out maybe overrated, just so that one, it’s done, it’s delivered, and also two, you’re able to just build again to that kind of.  

  

Consistent platform to a certain extent. And it’s definitely more cost effective to do it than to go back and do it again. Right? It’s certainly more cost effective to say hey, we’re building this core and shell we’re going to put four different fiber points of entry even though half the users only want two different points of entry, right?  

  

But we know these couple of other need four and we’re not going to go back and dig up the site and lay new pipe to pull in these extra fiber conduits. So in areas like that it’s definitely proactive and then honestly that over building just makes us a little more attractive because even though they’re standards too, like a lot of users, if they find out hey, I’ve got four poes, they’ll use four poes, right?  

  

Or they’ll use a higher floor loading so it removes some of the constraints they might have had that they put on certain projects due to financial reasons or speed reasons. We found it very beneficial to make sure that our designs are robust enough that they handle virtually any use case for these hyperscalers the biggest kind of deviation occurs at whether everything’s being stick built or the level of modular prefab building.  

  

That’s where it starts to get a little more complicated as certain users veer towards more prefab enclosure equipment in the yard versus others veer more towards stick building inside the building, more of a traditional construction style.  

  

So that’s honestly been the toughest challenge, is being able to accommodate both of those. Because if you’re looking for a user that’s looking for prefab and modular, we need bigger yards because they’re putting more equipment in it and smaller buildings.  

  

And then, you know, if you’re stick building it, you need a bigger building and and, you know, necessarily smaller yard. So that’s been our biggest challenge to be able to have the flexibility to go for both of those users.  

  

But to be completely frank with the state that the inventory is at now. A lot of users will shift their standard design, assuming the building meets what they need for redundancy requirements to be able to deploy that infrastructure.  

  

So when you look at Northern Virginia, when you look at Ashburn area, it’s less than a 1% vacancy rate in Ashburn. So there’s some real demands for the end users to become more flexible how they deploy as well, too, because they frankly just need to get these servers out there and working.  

  

Wes Edmiston: 

Yeah, I mean, touching on one thing real quick, something that I learned whenever we were talking the last time is so you said that there’s a 1% vacancy rate right now in Ashburn, so the demand is really high in that sort of Northern Virginia area.  

  

I didn’t realize just how many data centers there were in that region and how much data traffic goes through there. So can you tell everybody kind of like, what the stats are for data center construction and utilization and how much data flows through that area?  

  

Luke Kipfer: 

Yeah, it’s been a little outdated at this point, but there was a metric going around that 70% of the world’s internet traffic flows through two different counties in Northern Virginia. And while that might be a couple of years old, it goes to show just the amount of traffic that’s going through.  

  

I’m honestly going to get the number wrong, but it’s something like two plus gigawatts of just data center power in the Northern Virginia area under Dominion Electric. The growth is just amazing. So if you ever fly into Dulles, just look down and all those big boxes are data centers.  

  

And that’s honestly, there’s a lot of benefits as they cluster together, especially in terms of connectivity and fiber infrastructure. So there’s a lot of great benefits that come with. Having data centers together and you’ll find regionally, they do cluster together.  

  

You look at hotspots in the Dallas Market in the Phoenix Market in Santa Clara, Elk Grove in Chicago. There’s specific areas that once you get one or two of these big users coming in, everybody wants to come in and follow and be close to them.  

  

So Loudoun County was really ahead of its time in terms of tax incentives, in terms of ensuring that the necessary utility infrastructure was here and helping to promote data center growth. And that’s one of the reasons why they’ve started.  

  

But once one moves in, the fiber increases. The utility company sees it, they start bringing more power in, and then it explodes. It snowballs from there.  

Wes Edmiston:  

Even if that is an outdated number, I can’t imagine it’s decreased by too much.  

  

Even if we’re saying that 50% of the world’s data traffic goes through those two counties, that is monstrous. That is an astronomical number. So you guys are going in and you’re building this out. You’re working with these engineering firms and these other teams in order to ensure that you have everything up to snuff, right?  

  

You’ve covered all the bases for anybody who may want to come in there. I’m curious, what sort of trends are you all seeing as far as modular versus stick belt like you alluded to? What is the industry kind of moving toward?  

  

Are you saying more modular than before or maybe less? What can some of these owners do to help bridge the gap to where it’s not just you guys at powerhouse that are able to more effectively and efficiently build these things, but really the whole of the market, how can they better serve themselves?  

 Luke Kipfer:  

Well, as to bridge some of the gap between what some of these other owners are specifying to where they can get a higher quality, cheaper, faster product. Overall, definitely more of a move towards modular.  

  

And honestly, I think COVID had a bit of, I don’t want to say a negative impact on modular. But a lot of the choke points in modular have been the enclosures themselves and the actual packagers of these units.  

  

And the industry just got so stressed during COVID that what used to save a lot of time on modular for a while didn’t save that much more time because no one could get the parts. But I think people have seen, really the benefits of not just speed for modular construction, but cost and quality control, right?  

  

One of the things that I’ve seen that I think is underrated in terms of the benefits of modular is, frankly, like you coming from the trades, really, the safety aspect of it, and even the Ergonomics.  

  

So we were in a modular builder’s factory warehousing, and they’re building this huge piper act, and it’s something that’s going up in the air, but the guys are all standing there. They’re working on it on a normal working height.  

  

And then once it’s all done, they’ve got a crane. The crane lifts it up, they bolt it to the top of the enclosure, and it goes out, right? Versus if you’re building this in the field, some guy’s at the top of a ladder heads above his hand all day, trying to lift up pipe sticks on a scissors, lift it’s, those kind of things that if you don’t understand how they’re built, you don’t really see the advantages of it.  

  

But there’s some great things about building a power room. You’re inside, you’re air conditioned, you’re working at the right height, you’ve got all these quality controls. You’re not worried about dust, you’re not worried about the drywall guy behind you grinding and getting drywall dust.  

  

There’s so many benefits that come with this modular construction. So I think during COVID and through some of the as, the industry just didn’t have enough of these integrators that had enough of these enclosure manufacturers, that it slowed it down a little bit.  

  

But we’re certainly seeing a much bigger trend towards that, especially since, you know, kind of the dynamic nature of these data center projects is that if you’re able to standardize on a design, say, your Ups room, right.  

  

These guys can go out and they can buy these power blocks, these Ups rooms, and they can buy 50 of them and not know which site they’re going to. Right. They don’t have to worry about them being sent to Northern Virginia or being sent to Santa Clara.  

  

They can just give the PO to the enclosure manufacturer and say, hey, by the time they’re ready to ship, I’ll have a site ready. And they might be working on three or four sites at one time. And just whichever one hits first or whichever one they sign the client on or whichever one building permit comes through, they’re able to divert that as they need it to really meet their needs versus going out and pre building in Santa Clara, but then getting a client that wants to be in Northern Virginia.  

  

Right. So they’re able to really pivot and change how they deploy this infrastructure in a much more efficient manner than if they would if they were just going and stick building this. And certainly there’s different levels.  

  

Like, you can skid build it, right. You can put it on a skid and it can still go into the building and there’s different levels of it. But certainly every user is incorporating some sort of modularity.  

  

It’s just whether it’s kind of the single unit level or whether it’s getting to the point that they’re pre building entire electric rooms and shipping them around the country, the idea should be to get construction more likened to manufacturing.  

  

Right?  

 

Wes Edmiston:  

Yeah. I think modularization is a big way of doing it. But we still do this thing where, again, we try to specify the exact thing that we need for this one facility that might be slightly different than another that, you know, upgraded it a little bit.  

  

It could be a damn universal solution. Right? Yes. And we get so wrapped up in the exact specification that we miss the opportunity that’s there I agree that modularization is. Huge opportunity for that.  

  

Are you seeing more standardization though, I guess as time goes on to where we can take some of these benefits. Away from it, it’s funny. More standardization in terms of definitely design ups architectures, and there’s been a lot of standardization on that end.  

  

Luke Kipfer:  

The one thing that COVID introduced was lot of new equipment suppliers to the marketplace. So if you were buying switch gear for a data center five years ago, you really had three, maybe four players that you would go to since COVID since the supply chain issues, all of a sudden there’s ten guys, right?  

  

There’s a guy that there’s a lot of these small startup companies that saw a need created, went out, got the money, created their own manufacturing, and instantly got orders. So there’s new generator suppliers entering the market.  

  

There’s new electrical manufacturers, there’s new cooling, hot and cold aisle containment. There’s all these new players to the industry. So I think in general, the designs have standardized a bit. But it’s interesting that the specific manufacturers have really exploded in terms of who’s able to supply the market.  

  

But honestly, the way the industry is going and the way the industry is scaling is that we’re not going to keep up if we don’t standardize more. There’s no way that the industry is going to meet the demand if everyone’s off doing different things.  

  

And that’s back to how we design, is that having solid partners in the industry. For example, if doing a modular build, we’ll bring in that modular integrator at the very first stages of design because they might say, hey, I understand you’re specking this specific widget, but I got 40 of these widgets on the shelf.  

  

Just swap your widgets and I can get this to you 20 weeks earlier. So I think really it’s standardization, but it’s also bringing the trades in, bringing the suppliers in early, doing this really traditional kind of design bid build process just doesn’t work for moving quickly.  

  

So we’re at design meetings, we’re signing up electrical contractors early, sign up mechanical contractors early. We’ve got equipment suppliers in the meeting, and they all bring a different viewpoint and a different perspective.  

  

And frankly, they’re almost all always able to be able to speed up the project in one way or the other.  

 

Wes Edmiston:  

So you’re saying even outside of just, we’ll say hitting the same standardization, it’s also a willingness to be flexible and having those kind of open conversations with everybody, right?  

  

Like, hey, what can we do? Let’s all work together on getting this done.  

 

Luke Kipfer:  

Yeah, absolutely. It’s something that everybody could do a lot more of, just kind of in general anyway, right. It’s just not just talking, but listening.  

  

Yeah. No, everyone working in their own little silos. There’s no excuse for it. I mean, we’ve got the tech to involve people. We’ve seen the benefits of it. So it’s really breaking down some of these traditional and obviously it doesn’t apply just to the data center industry, but I would say across construction in general, breaking down these silos between design construction and the owner, the equipment suppliers really benefit everybody.  

  

Wes Edmiston: 

You bring up tech. I’m curious on what you’re seeing as far as obviously you were talking about tech as far as being able to communicate, but the thought in my mind is, what sort of tech are you seeing get introduced in, we’ll say data center construction to change the ways of working?  

  

Is there anything that you’re seeing as far as changes in new innovations and technologies that you’re seeing people deploy?  

 

Luke Kipfer:  

Yeah, I think we talked about it before, but originally BIM never really made it truly down to the subcontractor level.  

  

And if it did, it was really kind of the only the more advanced, like the nationwide electrical contractors would have a good BIM team. But when you get in the middle of nowhere, Bob’s Electric Service obviously isn’t running BIM, but it’s been so much more widely adopted.  

  

We were on a project a few months ago, big data center project, big electrical contractor. There wasn’t a single pipe vendor on site. I mean, 40, 50 electricians, not a pipe vendor. Right. They’ve got their BIM model.  

  

It’s all pre built in their warehouse. They stick it on trucks, they put a sticker on it, where it goes, what it’s labeled, what it goes to. Comes on site, electrician puts it in. It’s just amazing to me.  

  

Same thing like duct banks, right? We’re doing an outdoor duct bank. The hole is done, duct bank shows up, pre built, strapped machine drops it in, puts the next one in, right? Like, these guys aren’t in the dirt, in the ditch, putting together pipe.  

  

It’s all pre built. Shows up on a truck, they just connect the different segments and walk away. And it’s all pre built off of bit models. Right. So certainly that side of the industry, I think, is where it’s really been the biggest benefit that I’ve seen boots on the ground in terms of construction and tech and being able to move things forward.  

  

Another aspect that I don’t think it’s talked about as much, but the tech around safety, whether it be safety tracking, different tools. I’ve seen designs come out where they’ve got best guys wear that track their ergonomics.  

  

So if they stretch too much one way, or it’ll flag it to say, hey, you did the same movement 200 times in a half hour, that’s not safe. There’s some really cool ideas. I don’t think we’re there yet, but seeing some of the things that are coming down on the safety side of things is really great to see.  

  

I think the next five or ten years, in terms of what construction sites look like, it’s going to be vastly different, especially from a safety aspect.  

 

Wes Edmiston:  

Yeah, that brings up a good point. Even if. Even if any, I guess one of these solutions doesn’t end up becoming broadly adopted.  

  

It’s just the idea that once it gets introduced, it’s likely to get consolidated, incorporated in. As far as the general idea, once it hits the market, we can iterate on it, improve on it, and we’re at least trending in the right direction by giving the attention to safety that it deserves.  

  

I mean, even 10-15 years ago, while yes ergonomics were understood, nobody was going to track to see how many times you’ve made the same movement. Right? It’s just one of those if it hurts, rub some dirt on it, get back out there and get it done.  

  

But just the fact that people are taking the time, putting the energy toward checking the stuff, to me it’s encouraging. Right. Even the stuff where they’re tracking like body temperature to see who’s potentially going to overheat or trending towards something like a heat related issue, that’s phenomenal.  

  

I worked in South Texas for ten years, man. I know how it is. It’s a serious risk out there. So it’s great to see that the industry is moving toward it.   

Is there anything in particular that whenever you’re walking on a site that kind of catches your eye still? And are there any frequent issues, quality concerns that you’re seeing whenever you’re walking on a data center site?  

  

And what can we do to better improve these things? Specifically from the data center side?  

 

Luke Kipfer:  

Tough because especially as the industry grows, it’s moving into markets that have subcontractors that aren’t necessarily with installing data center specific equipment.  

  

So we really don’t have that problem when you look at an Ashford, when you look at in Northern Virginia, when you look at some of these more mature markets. But as the industry grows. We’re getting a lot of projects where subcontractors just aren’t used to working within data centers.  

  

So it was interesting, I was just talking to a consultant last week that he was on a job, and he said, you know, it’s a problem when the electrical contractor is reading the installation manual for the batteries, right?  

  

It’s just they’ve never seen it before, you know what I mean? Here’s this entire cabinet of lithium Ion batteries, and they’ve never installed them before. But he’s like, you don’t want the foreman in the field sitting there reading the owner’s manual before.  

  

I mean, you want them to read it, right? You want them to know what’s in it, but they’re doing it because they don’t know how to install it, right? So from our side, I would say it’s definitely looking at a lot of the data center specific items.  

  

There’s certain views on redundancy and quality that aren’t necessarily specific to data center, but specific to a couple of different sectors, whether it be data center, chip manufacturing, or healthcare.  

  

If these guys are coming from building office buildings and targets, they’re just not going to have that same perspective that we do. I would say, in general, across construction sites, I always look for neat, clean, organized construction sites, honestly.  

  

And it might just be in my head, but I haven’t had a lot of GCs go wrong when you’ve walked on, and you can tell they’ve got everything together and organized and neat, and they’ve got their subs in line.  

  

Frankly, a lot of the projects I’ve been on know, dirty equipment sort of mess. Subs are parking everywhere. You can really tell how good of control the GC has on the site by just the cleanliness, really, of the site.  

  

So that’s one thing that instantly, whenever I go by job sites or whenever I walk on job sites, I feel like I can get a feel for the project almost instantly. Just seeing what the signage is like, little things like that that go a really long way and.  

  

And then ultimately operationally from data center know as we’re building these ups rooms. And the data centers might not be to clean room level but they’re very, you know, really making sure that as a project develops that the subs understand that goes a long way to solving operational issues that occur later down the line.  

  

Wes Edmiston: 

Yeah as far as we’ll say the housekeeping and organization level and it’s one of those how you do anything is how you do everything. Yeah absolutely. Anytime that something goes awry, most of the time, you can track it back and you will find that the projects that are most messed up or where there were safety incidents, whether it’s the direct cause or not, you also had just a horrible workplace where everything was just chaotic, messy.  

  

There was no level of organization. And really it’s the condition creates a mindset which creates the probability that something is going to go awry.  

 

Luke Kipfer:  

Yeah absolutely. No 100% agree.  

 

Wes Edmiston:  

As far as people not being as familiar with necessarily data center specific construction, it’s funny you say the whole people reviewing the user manual for it or the installation manual.  

  

It’s one of those that you take for granted after a while of doing something over and over and over again, that, yeah, this is the way that we do things. Everybody knows how to do this but it’s not I mean for me in the piping world right, a pipe fitter is not just blanket universally a pipe fitter.  

  

They’re capable of doing this stuff. But the way that you do something in offshore is different the way that you do things in a chemical plant versus a nuke plant and everywhere else you go. So do you think that maybe kind of some of that opportunity in the technology space is toward upskilling people to be able to better onboard them to working in some of these, say more niche or unique working environments.  

 

Luke Kipfer:  

From my perspective, the tech aspect 100% helps. I really see it as a manpower issue. Right. These guys aren’t standing there reading it because they don’t want to or they’re not interested.  

  

They didn’t want to get into it before. But the GC can’t staff enough guys to train them. Right. The manufacturer is not out there because everyone is so short staff half. Right now that there’s a limited number of guys that know how to do this.  

  

There’s an even smaller number of guys who can train it. And frankly, a lot of the GCs that need to be supervising that guy don’t have the right number of people. And it just goes back to this growth.  

  

Right. The demand has so far exceeded our industry’s ability to staff it that we’re getting people from other industries, which is great and which is what we need, but we just don’t have the people there to be able to pass on the knowledge.  

  

So certainly on the tech side, how are we able to faster onboard to faster up level people to really set this in place? And nobody’s really come up with a great answer, right? Like the subs are kind of going one way, the GCs are doing something else.  

  

Manufacturers help with the training a little bit, owners help them a little bit, but there’s still a really big gap. And I’m sure it’s not just the data center industry, but it’s only going to get worse from here.  

  

And I mean, we could probably do a whole segment on skilled trades decreasing and pipe fitters retiring. And what’s the average age of the welder like 56, right? We’ll look at an operator, right? Yeah.  

  

Age of an operator is in their 60s nowadays. Yeah. So when you look at aspects like that, it is concerning. And ultimately I think it’s going to be the constraint on the industry, specifically a data center, but I’m sure other ones for how fast the industry as a whole can grow.  

  

Just because there is such a constraint on this skilled manpower that we’re coming up against.  

 

Wes Edmiston:  

Yeah, agree completely. It’s a concern. We’ve done a couple of episodes on that in the past. I encourage everybody to go up back there and listen to them, but everybody’s talking about it.  

  

It’s nothing unique. There is a major shortage, which is a great opportunity for people that are coming out of high school or college and maybe they don’t know what it is they’re wanting to do it. Not everybody has a dad that is as wise as yours, that kicks you in the butt and says, just go to work.  

  

But it’s real opportunity out there for people to make a good career for themselves and to grow and to mature to a position kind of like yourself.  

 

Luke Kipfer:  

Yeah. Honestly, that’s one reason why I’ve always frankly staying in Data Center.  

  

There’s just so many sub verticals right there’s. Construction, design, operations, finance, leasing, marketing you know what I mean? Data center has always been so interesting to me because there’s just so many different paths you can take and still be in this sector.  

  

Wes Edmiston:  

Yeah, I mean, something kind of tangentially related to this. I’m just kind of curious about what are you doing? So obviously you start off in the trades, you got your degree and then you got your Master’s, you got your PMP.  

  

You’re clearly the kind of person that is into continuous personal development. And like you’re saying, there’s a lot of potential avenues and opportunities for you to still grow in your career. I’m curious, what are you eyeing for your next thing in order to continue to kind of develop yourself?  

  

Are you going to go on and get your PhD? What are you thinking?  

 

Luke Kipfer:  

No, honestly, it’s this being able to network, to reach outside the industry, especially. I’m about two years in the Northern Virginia area and it’s really as you advance in your career, you see how important the connections and the network.  

  

Um, is to not just your development, but to your organizations. Right. Like, people want to work with people they like people want to work with people they’ve worked with before. People want to work with people they trust.  

  

Certainly I don’t have any sort of path outlined for formal education, but I have a lot of goals. There’s a lot of great data center, specific trade organizations, seven X 24, AFCOM infrastructure, Masons.  

  

There’s a lot of great groups that are really engaged in making the industry better. That’s one thing. I’ve really made a concerted focus to be able to spend more time with groups like that, to really start to see how I can help grow, support, maintain the industry, as well as meet people that are doing the same thing I am and figure out how we can do it better.  

  

Wes Edmiston: 

Yeah. That’s awesome. So sort of the networking and advocacy mean that’s one of the things that never ends seemingly. Right. So that’s awesome.  

Rapid Fire Questions

Wes Edmiston:  

Luke, we’re coming right up on time, so we’re going to roll into a few last minute rapid fire questions just to get to know Luke Kipfer the man.  

  

Not just Luke Kipfer, the professional. Right. So, Luke, what one word best describes you?  

 

Luke Kipfer:  

Honestly, I would have to say quiet. I’m naturally an introvert, so I feel like especially leading projects and things like that, it forces me to draw my inner extrovert.  

  

But the most times I prefer to listen than to talk. I know this podcast isn’t reflective of that.  

 

Wes Edmiston:  

Yeah, I wouldn’t have guessed that. That’s awesome. Generally, I’m more of a listener. Good job getting outside of your comfort zone.  

  

What is your idea of a perfect vacation?  

 

Luke Kipfer:  

It’s funny because I’m a project manager, right? I plan my life. I plan my day. I budget. You know what I mean? I do all the project management stuff for my life, so I try to idealize these vacations where I don’t do anything.  

  

And everyone I’ve tried to do that. I failed. Right. I’m going to go lay on the beach for three days and I just can’t do it. I make it like an hour in and then I start building an itinerary for the day.  

  

So at some point I’m going to find a way to be able to relax and just sit on a mountainside and drink a beer and not do anything. But I haven’t been able to get. There yet in the space of how to actually turn yourself off entirely.  

  

Wes Edmiston: 

You ever done one of those sensory deprivation tanks?  

 

Luke Kipfer:  

I have not.  

 

Wes Edmiston:  

Oh, try it. The first 20-30 minutes, you will be just reeling nonstop because I’m the same way. I cannot turn off. Then there’s like ten minutes that’s kind of kind of weird in between the last probably 15 minutes or so.  

  

Bliss. I mean, it’s off. I love it. I highly recommend it. What’s your favorite book? 

 

Luke Kipfer: 

I got to go with how to win friends and influence people. Dale Carnegie. Love it. That’s a classic. But that’s one I always go back to.  

  

Wes Edmiston: 

Yeah. What is your favorite quote?  

 

Luke Kipfer:  

I do, and I worked with a guy on the hyperscale side that loved this one. That was a smooth sea. Never made for a skilled sailor. I think projects always have issues.  

  

The job doesn’t get easier, you get better. So I think there’s always problems, there’s always issues, there’s always things to work through, but I think at the end of the day, they make us better people by figuring out how to manage those.  

  

Wes Edmiston: 

 

You were spot on with that. What’s your dream job?  

 

Luke Kipfer: 

 

I’m doing it.  

 

Wes Edmiston:  

This is it?  

 

Luke Kipfer:  

Yeah. I’ve got hobbies and things that I would like to do, but I’m lucky enough that I think I’m to the point in my life I can pick what I want to do and I wouldn’t pick to be anywhere else right now.  

  

Wes Edmiston: 

That is awesome. And if you could have dinner with any one famous person, living or dead, who would it be?  

 

Luke Kipfer:  

That’s a tough one. And I’m going to cheat. I’m going to say dead is Abraham Lincoln. I really enjoy kind of his take on life and in such a stressful, hard time.  

  

How he managed kind of his reactions and his personality and how he made decisions, I think is amazing. You know, for a living. I am a big podcast nerd, but I do like Andrew Huberman. I don’t know if you.  

 

Wes Edmiston:  

Oh, love it. Huberman Lab is fantastic. My favorite podcast.  

 

Luke Kipfer:  

Honestly, I would love to just sit down and pick that guy’s brain for like 2 hours. Right. That would probably go a long way with my own health and fitness.  

  

Wes Edmiston: 

Yeah, I know. Agree with that completely. Also, I’d like to just go to the gym with him and with did you listen to the series with Dr. Galpin? 

 

Luke Kipfer: 

 Oh, yeah, yeah, no, that was great. I don’t know what it is, but there’s something about Dr.  

  

Andy Galpin that is just highly familiar to me. Like, he just seems like one of my friends, like people that I grew up with and I just love to go to the gym with the guy. That’s what I want to do. So yeah, have dinner with Huberman and just go to the gym with Galvin.  

  

Wes Edmiston: 

That’s to me like a perfect day. But that is awesome. Luke, this has been a great conversation. I really enjoyed it and I hope that we can do this again sometime.  

Facebook
Twitter
LinkedIn