What We Learned About Point Clouds (feat. Matthew Byrd)

Listen now or check us out on one of these platforms!

Apple Podcasts


Google Podcasts



This week we talked with Matt Byrd. We had the pleasure of working with Matt and the Nexus team on a project, and wanted to walk through the job we worked on together. 

What are some common misconceptions of point clouds? 

How do the trades use point clouds differently? 

We talk about how to use point clouds appropriately, when point clouds work, and even our own personal pinch points in using point clouds. Laser scanning is cool, but let's talk about implementation and use cases.


Matt’s LinkedIn

Nexus Website

The Breakers Hotel in Long Beach (the project we worked on)


Tyler: Welcome to the Construction Brothers Podcast. I'm your host, Tyler Campbell, and with me like he is every week: My brother, Eddie Campbell. 

Eddie: What's up, Tyler? 

Tyler: Not much, Eddie. Well, this week: Laser scanning. What did we learn?


Interview: (12:22)

Eddie: All right, Matt, thanks for joining us today, man. Why don't you tell us who you are and what you do?

Matthew Byrd: Awesome. Thanks for having me, I appreciate it. So my name is Matthew. I run Nexus 3D Consulting and I also host a couple podcasts: The Reality Capture Network and The Apex Mindset Podcast. But mostly in the construction world, which is where we kind of met you guys from, is with Nexus 3D. We are a geospatial reality capture services company. We do conventional land surveying in California, that's where we have our license services. But one of our big focuses is helping companies with laser scanning, point clouds, UAVs, and how to apply new technologies into different construction processes, and really through a lot of industries. We work a lot in industrial, oil and gas, food processing, and then we get into architecture, BIM, airports. We've really found that these technologies apply through a lot of different industries, really around the same issues, right? If you're going to renovate a building, you're going to expand on a building, you have a complex facility and you need to replace equipment, you need to take things in and out—It's hard for people to do that without accurate information of what exists there before they start design. And we know in the construction industry, there's a lot of waste, there's a lot of issues. And so looking at these new technologies that have a way to reduce that and help the design process, help fabrication, help installation, that's really where our focus has been, is around the technology and how it can apply to different industries. So that's a little bit about what we do and I'm excited to get on here and talk with you guys today. 

Eddie: Excellent. We crossed paths because we were involved the 3D modeling, structural steel detailing, BIM side of things on a project called The Breakers historic hotel out in Long Beach. And you guys did your whole spectrum of services on that, with scanning and helping the main client who is ARCO/Murray out there, figure out what they actually had in this old building. So I want to hear some of the things that you used, like what exactly did you do to gather information on that building? 

Matthew Byrd: Yeah, so this project was one, like many, where we weren't involved from the very beginning of the job. We didn't really know when things got started, they didn't already know, “Hey, we need to scan this before we ever get into design,” the project was already moving along. They found a reason to look into scanning. So it was new to the people on site. There were already firms out there doing survey control. We do our own control as well for projects. So really, we got pulled in and our initial focus was capturing the existing conditions of the building. This was a 13 story building with a basement, so the exterior was quite large, which also brings in some complication of how do you capture those facades on the roof? So what we ended up doing as far as capturing the data was we had survey control set. The building is going through renovation. So there's windows, there's openings, which is nice because you can get survey shot into each floor. And that helps with some of the accuracies of tying down data per floor. So what we did was we had survey control on the exterior of the building, as well as on floors inside of the building. We used terrestrial laser scanning, which is the tripod based. It sits there and spins and uses a laser to collect measurements as well as take images from different positions. We use the terrestrial scanner around the outside of the building on the site to capture the facade from ground level. We also made our way through the interior floors mapping each floor. And again, this building was in the middle of renovation. So it was gutted. It was mostly just visible concrete and structure, which is nice ‘cause some of those floors were pretty open, and we made our way throughout all the spaces through the stairwells and using that scanner to capture all that existing data and then tie all that together. So what you end up with is a full mess of millions or billions of points that have been measured, you know, XYZ coordinates of that entire structure and the site surrounding. Something else we incorporated on this project, which was unique, is we used Aerial LiDAR as well. You know, the building again was super tall. It was hard to get roof and some of the exterior facade. So on top of doing the terrestrial scanning, we had Aerial LiDAR flown and captured the roof and the outside facade. And we married the two of those together with the controls. So we ended up with this point cloud that was made from the terrestrial and the Aerial LiDAR. And then all of that tied together to one point cloud.

Tyler: I have to say, too, after seeing this all put together, I mean, that was a pretty killer looking point cloud. That was pretty impressive. How accurate are these laser scanners that you're using?

Matthew Byrd: So this is one thing that comes up a lot. How do you define the accuracy? A lot of people find out about laser scanning, they find some random unit there. You know, you hear the specs of what the laser scanner says, right? The laser scanner is one millimeter accuracy. Well, that is a given measurement from the position of the scanner. And then you have this error that gets built in when you're going through a project where, you know, if we're on a single floor within that project and we do 15 scans or 20 or 30 or 40 scans, we have to tie all those together like a puzzle. And by the time you tie 40 scans together, you're not keeping that one millimeter accuracy because there are targeting, there are things you have to do to tie those scans together, and depending on how you control them is going to change what your overall actual accuracy is within that floor. That's just one floor. By the time you go through stairwells and you go through and you're doing, you know, a couple thousand scans, you're having error that can be built up between those scans. Now this is one reason why survey control is important, and a lot of companies don't know that. There are a lot of people who learn about a scanner and its way to collect data and how they can use it, but they don't understand conventional survey. If you try to do a project this size without survey control, you're going to have inches, lots, lots of error. By the time you get through 13 floors inside, outside, basement, you're going to have who knows what. You know, six inches, a foot, like there's a lot of ways that things can go wrong if you don't incorporate survey control. And you also can't prove it. If you register through—and I like to use this as an example—so whether it's a elevated or a linear project, you know, if you're going through stairwells and you're doing two floors at a time or something, you know, say that you do the registration process and there's an eighth of an inch error, right? Between those floors. Well, by the time you add on an eighth of an inch through 15 levels, what's your overall, you know, what are you really at from top to bottom now? Or same on a linear project. If you're going a thousand feet of length and there's, you know, a millimeter or two millimeters or three or four or five millimeters of error between every single scan or between, you know, a hundred feet at a time, you have a lot of error built up. What conventional survey control can do for you is it can give you known coordinates on the top level, on the bottom level, on the exterior. And the survey control is a very, very accurate measurement. It is held to that finer, you know, eighth of an inch is what we typically go to on the survey control or maybe, and that depends too on the process that's used, the accuracy that's needed for the job. All of those things are very important when you're talking about that. So the scanner itself, if you do a single scan and you're going to get really, really tight measurements, but then the process on how you control those laser scans with spheres or checkerboards or survey control is really going to be what determines your overall accuracy for a given area or for an entire project site.

Eddie: I wanna hit the why a little bit, like why did we bother with a laser scan? This building, I mean, just to describe it a little bit if you haven't heard of The Breakers on Long Beach, it was built in 1926 and it kind of is overlooking the Pacific Ocean. It's a very historic building. Had some pretty cool movie stars come through, had a really notable restaurant on top. Like this was the place to be once upon a time. One of my favorite things that was retained through the scans was the gun turret that was on the backside of the tower that looked out at the Pacific Ocean from World War II times, which was just a really cool thing. But the building was beat up. I mean, there were places where the concrete was worn so thin that you could see through to the next floor. You had a lot going on in this structure after demo. But we're trying to restore this building, and there was a lot of new steel and construction that needed to be put into it. So that's how our two companies worked together, right, so we were taking those, the scans, the scan data, the modeling data that you guys had compiled and trying to get the building to fit with very non-malleable material and steel, just trying to get the thing to come together as easily as possible. So that was no small task, and definitely a need. So I wanted to kind of turn the page, like, how'd you think it went? 

Matthew Byrd: One thing that we noticed on a lot of projects with these new technologies is, like I mentioned, it wasn't brought in at the very beginning of the project. All the team members weren't pulled together, even. When you're looking at integrating 3D data, UAV, point clouds, laser scanning, drones, 3D models, as-builts—you have to also know who all is going to be working with that data. What softwares are they working in? Can it take a point cloud? Can it take a Revit model? What file format? There's so many processes that have to be walked through in order for different trades to be able to use the data properly. You know, if you have one design team that's not even working in 3D, how beneficial is the 3D data going to be to them? You know, how do we then have to dumb that back down into 2D for somebody in 2D to use it while somebody else is working in 3D? I think the best way to apply the technology is by us continuing to educate people, talk about it, and then for teams to realize, okay, the best way to do this is talk about it at the very beginning of the project. So that these questions can be gone through, because I think that's something we ran into on this project. We got pulled in to come scan, but we didn't even know who all was using it, what trades, what softwares. And so if you kind of get piecemealed information and piecemeal deliverable requests, you aren't really serving all of the people that need that data. And that was one of the things I think that happened is, you know, if we know that an architect or somebody is working in one platform, then that's the process we have to follow to create the deliverable. And there's different accuracies, too, right? A lot of times an architect is asking for less accurate data than maybe someone who's prefabricating steel, and maybe the processes used to take a point cloud and create the as-built model that that architect needs is completely different from what you might need to do for the dimensions from the steel. So that's why the team discussion around the deliverables is so important. The data is, is really nice, really accurate, really usable, but it's also still in an early enough stage that, you know, a lot of people think you just hit this button and the point cloud, you can just produce out drawings and produce out models. No, it's a very manual process. And there are different softwares out there to help extract data and to help create models from the data, some of them are faster than others, some are more accurate than others, some create elements directly intelligently within a software platform, some are dumb just CAD models. So there's a lot of directions the data can go, but it has to really be driven by and dependent on who's going to use it, what platform, what accuracy need, and I think that's something that you guys realized as well. And again, this is the technology still being applied. There are a ton of firms who don't even know laser scanning exists. They've never seen a point cloud, you know, they're out there with tape measures. So when you get into this technology for the first time on the job, or even the first couple of times, there's learning curves. And again, especially when you're including different trades, because a lot of people are working in different platforms and it all has to figure out how to communicate together.

Eddie: Yeah. The process got interesting. We made a site visit January before last. That's how long this project's been going on. And this discussion arises in the Uber on the way over to the job site about how we accurately model to what we've been handed by our customer. And so we've got a Revit model, and we've got scan data, and then we've got CDs, and we're trying to figure out, well, what governs? What's the intent, what are we supposed to go to? And you know, we've got a 2D architectural effort. We got a 3D structural engineer effort. We've got a Revit model, we've got scan files. Our companies had kind of worked in vacuums apart from one another and had not really been introduced. And then on top of that, here we come with the Tekla Structures, structural detailing that we're trying to work to all of that.

And so, yeah, and the expectation was we'll be able to produce shop drawings to just hand to a fabricator and the fabricator can just go. Well, it was interesting one thing that happened, there was a beautiful Revit model, beautiful rendering of this building, this historic building, a lot of work went into that. We were pretty immediately told, like you can reference only, but don't do anything with it. And then the next thing that happened was the fabricator was bound to actually field dimension everything out there to make the steel fit. And so that didn't, to me, feel like we had got all of the value out of this process. I don't know how you feel about that.

Matthew Byrd: Yeah, no, I agree. And kind of like I said, when you know what all trades are working with the files and even within— We'll talk about just Revit, you know, and I'm not personally the Revit guru, but I work on these projects and I manage them overall. Even within Revit you can have models that are built perfect 90 degree angles but best fit, which is what a lot of architects want, but that means that you're not going to have the exact accuracies and dimensions if you're trying to get really fine measurements for steel. You know, if you build a best-fit 90 degree beam or column, but the column’s actually leaning and angled—well, if it's best-fit and you put it in the middle, then in the middle it's going to be right on, but at the top and the bottom, it might be off. When you get into an as-built project with scanning, the scanner captures every defect, every angle, every deformation. If you're going to take that point cloud that has the surface mapped completely accurate to how it actually sits, and you're going to turn it back into a 3D model, where do you place those elements? And that's something that comes up on a lot of projects. Some clients, if it's just an architect using a floor plan, they're like, no, we don't want to see these angles in the walls. We just want it 90 degree angles, we want it best-fit in the middle. It really is going to be dependent on who's using it and their needs for the project specifically. 

Tyler: I think that's a great example of basically just how effective scan data can be. I was recently on a project, not Breakers, but on another one that was a retrofit. And they were trying to install this steel mezzanine in this building for a coffee bar. Well, they have all these concrete columns that are in there and they're twisted, right? So you're talking about going in and measuring. They just smack the measuring tape up, you know, chest high, or down on the floor, but they don't take into account that as it deforms over 14 feet where they're actually going to be installing the steel at, yeah, you can't take any of that—

Matthew Byrd: It’s in a totally different location. 

Tyler: It is. Yeah. So I could be half inch, an inch off, which for steel—I mean, you can accommodate that, but in a lot of cases, the engineer doesn't actually allow for that. They don't give us the leeway that we need for slotted holes or things along those lines. So it can be so effective. I think that's one of the biggest gaps in applying this technology right now is this discussion—is that somebody scans it, somebody hands over a point cloud, they don't know how to use the point cloud. They have someone else create a model, but is the model built the way they needed it to be built? Or is it even built in the right platform? Are they as-built modeling it to the actual fit of the data? You know, because how do you go in Revit or in one of these softwares and model out that beam you're talking about that’s got this twist and weird angle? And, you know, it's really not meant for that. We're taking this software that's built, it started from using it as a design tool, and we're trying to then incorporate this really complex 3D point cloud, which it does bring it in, but it's not a software that is built to just perfectly map out these really complex figures. Now there are ways to model that stuff, but you're talking about, okay, do we best fit a column or a beam and angle it to just be best-fit? And that's like a couple clicks of a button. Or do you actually need a physical 3D model that is fit to that surface? Well, now we're going to have to go to some other software that lets us do that complex model and then bring it back in because you're using Revit and now it's not a family that you needed. You know, there's so many different things that come into play in that process. I think ultimately, like you said, the laser scanner is amazing. You know, it maps every surface. If you can figure out how to utilize that data and measure what you need or bring it in, or you know the model that needs to be created and somebody can do that process for you, then there are times people want or need the model created. That's another big discussion right now, right? Why don't we just use the point cloud then? Well, this point cloud was hundreds of gigabytes of data. Most design firms are not prepared to handle that. Their computers hardly have enough space to hold it. You know, they don't run opening that, you know, it’s bogged down or some of the software they're designing in can't even bring in the point cloud, or if so, it's way too big. There's so many complex issues with the point cloud itself still. I think that's going to continue to be improved and developed. But ultimately the biggest piece is, okay, the scan is done. It's really awesome, it's really accurate, but how do we use it properly dependent on each trade or what the project need is or type? And that's why I think the discussions need to happen up front, because if a construction company just calls us and says, “We need you to scan this building,” well, okay, we can do that. But then you miss all these questions, and the construction company doesn't know the answers to these questions either. If they're our contact and we talk to somebody and they're like, “Yeah, we just need you to come scan it, give us the data.” And we're like, “Okay, give us what data? The point cloud or you need a model created?” “Well, I don't know.” “Okay.” “Yeah, it sounds like we need a model.” “Okay, Revit?” ‘Yes Revit.” “Okay. Well, that's what we did.” You know, you really got to get all these parties involved to have these discussions. What's important? Why are we scanning in the first place? What do you need to be able to measure out of it? Can you do it from the point cloud and imagery, or do you need a model? What trades are taking it? What softwares are they using? Can we create a model that can be done that meets all of those needs, or do we have to create separate deliverables for each trade? There's a lot of things to look at.

Tyler: Do you run into an issue with trust from your work? Like, have you ever had somebody push back against you and say, “I don't trust this point cloud,” or even the model, like a Revit model or anything like that, that you've created?

Matthew Byrd: Sure. Yeah, that does come up. I mean, we've honestly had people that have told us, you know, we scanned a building— Again, we know the accuracy of the data that we scan. We register it together. Even on a small project, we had an architectural building that we scanned, and it showed that the walls of this building are like 88 degrees when you come to the corner and we've had people say, “There's no way this is proper, I've been to the site, I've been in design for so many years, there's no way that that was built at an 88 degree angle. These surveyors lay these column lines out perfectly.” And we're like, “I'm sorry, but the point cloud of this building is showing this angle. That is a hundred percent how it sits out there right now.” Some people just don't even believe it, because in their mind they think that this building is going to have this perfect angle. And we're like, I'm not even in construction and I know that's not true, because we use these technologies that map it right where it sits. And it might've been designed to be a perfect 90 degree angle. And even in the survey world, we know surveying, we know there's tolerances when you're out there staking out columns to be set and then the construction company puts those in. Those are not perfectly aligned. If we model out where those columns actually sit and draw a perfect line between, some of them are up, some of them are down, but some people don't know that. They think that it's going to be this perfect laid out system and it's not. And the scanner is the best way to show that. And then the next step would be the model. And again, that really comes into, what's the need, what's the accuracy and how we're going to go about modeling it. But there are definitely people that don't trust the technology, and there are things that can go wrong. One thing I always try to recommend to people is if you're getting into scanning on a project, make sure you discuss with someone that at least understands the survey control side along with the scanning side. Even if they're not the ones doing the work, but at least discuss it, because it's way better to find those things out and plan ahead, then scan a project that, you know, you register together, you didn't know about survey, and now you do have this error built in and it causes issues. That happens a lot.

Eddie: Tyler, how many times do we boil back to this same kind of argumentation of planning?

Tyler: Oh, every single week. 

Eddie: Doesn't it feel like we keep coming back here? I was thinking about the “measure twice and cut once” thing. It's like, no, before you commit to that thing, measure it one more time. And then, you know, let's be sure before we act, think before you act. For this particular project, we didn't deal with a distrust of the information. Rather, we had an innate trust that almost made this like a magic pill. Like, what about the scan? Like, the scan took care of that, the scan took care of that, there's no need for anything outside of that. But the process actually looked much different than that by the time the fabricator got done doing his field dimensioning. That was interesting. Just being on the side of having somebody to just tell you, “Yeah, you're gonna use the scan, everything's going to match up and it's going to be great.” And then realizing that in the particular version of the software we were using, we couldn't even snap to the point cloud. And I'm going, well, how do I know where that thing is when I can't even snap to it? The best I can do is get in a section, look at the plane, and try to figure out where it is. Another thing, I mean, we're going out to the site and we realize what you were talking about, a standard concrete beam does have a taper because that's how we break the forms. Well, all of the beams were tapered and they were tapered about an inch on either side. They were flat in the Revit model because we went for best-fit, which made sense for the application. But the reality of the connection and then what the structural engineer took from that best-fit model were two different things, because the best-fit model showed a square beam with a nice orthogonal connection. None of the connections work that way because they were tapering away from any beam that was connecting into them and it caused a huge issue, because now, just standard connections don't work anymore. I mean, just the one inch on every single beam that didn't get represented completely changed how we connected everything.

Matthew Byrd: That's what we're talking about, about knowing who needs it, how does it need to be modeled? Because to be honest, this building got modeled a couple times. It got modeled to where every beam did have proper taper and it was custom to the exact beam. But then we were told, you know, “Hey, this isn't working the way we needed it to in Revit, it's not showing up properly when we produce these sections and stuff.” And we're like, okay, but that's because we fit it properly to exactly how it fits. So then we had to remodel it to the best-fit. Then our question is like, okay, well, if we're modeling it to best-fit, but you want to use these certain elements, do we include this angle that's on there or no? And it's like, well, yes or no? Now if one person makes that decision, but you don't know about it, or you don't know about it, or this other trade doesn't know about it, they think that that's actually how it is in the field, but it's not. There’s so many of these processes, and I think that there is a path forward, but you need to have everybody involved so that at least you guys, the architect, the construction, the structural, the mechanic, everybody knows: What's the point cloud? How accurate is it? Can we use it? What's the model that's created from it? How was it built? Is it best-fit? Is it proper? You at least have to know, because again, I think there is a solution. If you need those taperings or you need a certain column or things custom-built, it can be modeled that way. A hundred percent, no question. Yes. That could be handed over to you exactly how you need it. But if that doesn't meet the need of the structural or the architect and they want to build a different way, well, you're almost having to create three or four different models. And that's okay, too, but then it's going to cost more. Who's paying for it? Are they okay? You know, you got to think about all of that in whole. And like you said, too, you come back to this planning stage and you come back to, how is this going to help us? You got to talk about, again, who's paying this, who knows the difference between why create a model or not, or maybe creating two models or three models or whatever it is that's needed. Who is it going to help? What potential risk, or what risk mitigation, or what savings in time or error or whatever, who's it going to help? Because the cost of doing that scan and creating the model, maybe even creating two or three or however many models is probably worth the headache to avoid all the issues of these different trades, trying to deal with something that's frustrating and it's not working for their needs and their platform. But you've got to have the discussions.

Eddie: There’s a few things that there are problems with, but I don’t want to keep calling out the negative parts of this process. We're looking at a project and doing the postmortem almost like in front of everybody. Like, “Hey, look at this,” because we want to be authentic about the fact that this is awesome. This was a good idea. We're glad they took this path, but now let's look at it and see how it could have been improved. So not just like slinging muck and just saying, “Well, everybody was really dumb about the way they did it.” No. But next time, what would we do differently? How might we adjust? How might this go better? And so I see issues with big data. I know the first time we ever got a point cloud, we were asking how we were going to get this and thinking, “Yeah, Dropbox it over to me.”

Matthew Byrd: Yeah. It'll be there in a couple days.

Eddie: We got a hard drive in the mail, for real. 

Matthew Byrd: Yeah. We still get them in the mail.

Eddie: Because that was honest to goodness the most logical way of sending it. The way our particular software houses it, there's a need to have the actual scan data live on the C drive of the machine. So you think, if I've got five, six projects and I got 200 gigs a piece, I'm running out of hard drive space in a real hurry.

Matthew Byrd: Yeah. That project alone on our end, in scan data and conversion and stuff, was over a terabyte. You know, I think it was maybe one and a half terabytes or something like that. And when you're trying to then hand that over to people that are not used to this data, it's impossible. And again, that's the importance of this is like, there's ways to still apply it. Like we can reduce down file sizes. We could give a floor at a time. And that's one of the big reasons for a model as well, is these architectural firms, they're not ready. They don't want to deal with that point cloud. It's way too big. They don't have the capacity for it. You know, they aren't familiar, it's not in Revit and they can click on it and produce a floor plan like they're used to—you know, there's a reason to have the model, as long as it's done the way they need it to be done. But the point cloud is very usable, too. And I think we're just, we're in that phase that people are starting to learn about it, how to use it. It does take upgrading machines because most people don't even have room to open a project that size, especially on their internal drive. So data is a big, big issue within the technology still.

Eddie: The data side of this, this is like the data confusion side of things. So you're talking, like, what do I need? How do I need it? And for whom do I need that? And let's say that ends up in a project where the owner or contractor says, “Yeah, I'll bite it off for three different models for three different uses.” And I got three different models living out there for three different purposes. And I mean, so how do I federate that? How do I control that data? How do I make sure it stays in the right hands? And then how do I update all three cost-effectively? That's an issue.

Tyler: That's where the Autodesks and the Trimbles and all of them are trying to step in and come up with solutions so that it can all be passed down. But honestly, that's just a version of Dropbox or Drive, you know?

Eddie: You know what’s funny is that Tekla—we were in 2018i for this project, and it didn't bring in a snappable point cloud in 2018i—in 2020 now, you can actually look at collisions with a point cloud and it will give you a gradient of collision. So just in two versions of this software, it's obvious: Autodesk, Trimble, Bentley, they are all looking at this going, “Hey, we got to get on board because this is—“

Matthew Byrd: Oh yeah. I mean, some of them are— Autodesk has been one of the big pushers of it, right? I mean, they have integrated directly bringing point clouds into all their platforms, into civil 3D for our survey work, into Revit for the architecture stuff, into 3ds Max and entertainment. So it is continuing to be developed. There are individual software companies, apart from these big guys, that are working on tools to better model from it, better integrate it, better analyze it. Everybody is going this direction. And I think we're going to just continue to see that over the next couple years, but it's awesome to be on that front end and be with people that are first starting to integrate it, because we collectively as the industry are the ones directing the software and the hardware on what's working, what's not. “Hey, Tekla, we need you to be able to snap to point clouds,” boom, it's in there now. Like, what's next? What else do we need to work on? How do we continue to discuss these pros and cons of using these new technologies on our project between different trades, and then direct Autodesk to, “Hey, here's this issue, how can you get us this model that’s going to serve all these trades, and then convert it back in to be able to be one model and federated and coordinated and whatever?” It's going to continue to get there.

Tyler: Well, man, I think this is a great place to break off and ask you our megaphone question. So if we gave you a megaphone that the whole construction industry could hear for 60 seconds, what would you say?

Matthew Byrd: Yeah, I think I would probably say, along the lines of what we've discussed, is look at new technologies and plan ahead with your entire team. You know, a lot of times, even in construction, you get the guys on site, you get the superintendents, you get the people that are pushing, pushing, pushing. “We need this today. We need this tomorrow.” It's like, if you breathe for a second and look at these new technologies that are out there, get everybody involved in how to use them, I think you're going to see that, even if you delay the beginning of your project a little bit and do this planning, you're going to turn around and see that that was worth it. Because you didn't run into these issues that some of these projects do, where there was a break in communication, certain people didn't even know what was going on—if you do that up front and implement these technologies, implement the better communication and planning, you're going to see that the rest of your project is going to be on fast forward, because you took that break in the beginning. 

Eddie: Word.

Tyler: Definitely. Well, Matt, where can people find you? 

Matthew Byrd: Well, let's see. For Nexus 3D, our website is nexus3d.com. One of our most active platforms, we’re all over LinkedIn. I personally am trying to really do what I can to continue to spread and educate people on these technologies, which is why we started the Reality Capture Network, which is also there. And we're on all the platforms. You can find us on Instagram, LinkedIn and our websites. I appreciate you guys inviting me on, and we look forward to continuing to push and innovate alongside you guys and others in the industry. 

Tyler: Heck yeah, man. Appreciate you being here.


Tyler: Hey guys, thanks for joining us today. I wanted to take a second and point you at a couple of things before you go. Number one, make sure that you go check out our website, it is www.brospodcast.com. We're constantly posting new blog updates on there and thoughts of things that we're seeing in the industry. And then also, make sure while you're there, go check us out on social media. We are on Facebook, LinkedIn, Twitter, and Instagram. We're constantly asking questions there, trying to get people involved and engaged and learn more about what's happening in the industry so we can keep bringing stuff to you that’s relevant. And also, share it with somebody. If there's something in here that you thought was valuable, forward it over to your friend, let them know about the show. Again, that's going to help us out a lot. And finally, please leave a review. If you found this interesting or helpful at all, you could help us out in a big way by just hitting a review on Apple Podcasts or Stitcher. So thank you so much for joining us this week. Have a good one.