Bản chép lời đầy đủ của How Technology Is Reshaping The Construction Workforce with Dexter Bachelder

trades rather with technology or robots, as you would put it. I mean, my personal take on this is is you have to remove the emotion from the reality of the situation, which is we need half a million people, right? We like there's like that's like jobs being automated or not is beside the point. The reality of it is we're short what we need. So even just like digging in on you mentioned robots with like drywall, right? And this, what do you what do you think is like the reality of us getting technology to do an adequate, if not advanced, job in these trades, yeah, I think it's I think it's a challenge, and I don't know if robots are going to solve the problem over the next kind of you know 12 or you know 18 months, and you know AI can automate a lot of processes. I think that there's always enough work to do in a project site, that if you know there's more than enough work for if someone automates one part of somebody's job, they can go off and do something else, or they can work on two projects. And so I just need that that excess kind of capacity and and leveraging the tools to kind of you know also get people home at night, like and get home safely at night, and instead of you know staying on the site until 8o'clock at night, going through you know logs or something like that, you know let's let the tools do some of those things and improve the quality of life and and the safety of of people out in the field. 00:06:54.339 --> 00:07:41.740 Totally, totally, and like even just like personally, like I've got I've got a family member I was talking to him on Sunday, and they have a huge construction project going on, and all the the project management team and the superintendent team they're doing shifts over weekends just to like keep eyes on projects and people doing what they need to. So, like the need is certainly there, but I think the industry is still you know skeptical, as I'd mentioned earlier, of like what is the validity of this actually happening, right? So I maybe like from your point of view, how how do you think about or how would you advise contractors to think about finding areas that they could apply technology, whether that's an AI solution or robotics with AI, right? How would you urge them, or how do you think about applicable places to start automating these workflows. 00:07:42.819 --> 00:07:56.199 Yeah, I think you know one one thing that I'd like to cover too is that I think we need to be in tech-enabled industry to attract new workers to to the industry. 00:07:50.649 --> 00:09:44.590 You know, we have tools, we have iPads, we have apps. You know, we have 3d models. We we can attract more people, you know, to leave other industries to come into the construction industry to be in a tech enabled kind of industry. So I think that's one area that a lot of you know companies need to do is kind of show some like cool innovative technology to you know I don't know teenagers like games and a three model is not 100% different than a game, or you know some of those kinds of things. So let's let's make let's use the technology to also be more efficient, but also attract people into the industry. I think go back to your question. I think you could look at where you know do good research across all the kinds of different solutions. And you know, when I started 25 years ago, there were a handful of companies, and now we have hundreds or 1000s of kind of point solutions. But where where's the real pain? Where are the delays? Where are people struggling? Where are the bottlenecks? You know, what's taking longer than it should? I think those are the kind of areas that you got to start to identify, and you know I you know people have innovation teams and technology teams and but but I still really feel like it's the you know people on the front lines that need the tools and have the pain and and finding solutions to their pain is kind of you know I think that's how we we get more adoption and usage and that's how we kind of start so, and I've never seen a technology rolled out where if you didn't have the people in the front line embracing it, you know, it it won't be successful. And you want you know Joe on one project to say oh you know to Mark on another project hey we're using this tool and it's you know it's been awesome it's you know saved me five hours a week, you know, and those are the kinds of things that you need to drive adoption. 00:09:45.460 --> 00:10:12.059 Totally okay. Well, and so I think your point with like being able to attract talent by becoming and continuing to lean into, which I do think the construction vertical has been doing a great job of over the last call it five years, but really becoming like a forward thinking. Tech-enabled industry, I definitely think that's a solid point. Will tie into that, but even in your perspective, because you nailed it, right? At the front lines, if they're not adopting the solution, they're probably not seeing the value. 00:10:10.139 --> 00:10:45.250 If they're not seeing the value, then the solution is not going to stick, right? But you also mentioned like going in and evaluating what are the workflows, what are the processes, what are the areas for improvement that we could start applying solutions to, and just from my own point of view, being on the software solution side, is like that is really easy for someone like me to say, especially when we're focused on construction accounting or Speckle is focused on BIM data, right? Like it's really easy for us to understand. We'll just focus on these internal processes and then start evaluating products from there. 00:10:45.850 --> 00:11:04.620 But is there any more depth that you could advise a construction company to start? Like, what are some like tactical, like one or two steps that they could start with in terms of importance to be like, okay, operations from this standpoint, that's good to go. Back office is bleeding. 00:11:01.169 --> 00:11:16.740 Maybe we need to like

look at these triggers and evaluate these processes. Like, how would you urge biased aside contractors to start really looking at their processes internally and then become decisive on what's the most important workflow to start automating? 00:11:18.299 --> 00:11:22.429 You know something that I've seen a lot over the last 20-five years. 00:11:23.059 --> 00:11:34.279 Yeah, is that where you've had two champions on on either rolling out a technology or a project? 00:11:29.539 --> 00:12:50.710 You kind of have maybe, and doesn't have to be, but maybe a younger person that's kind of really tech friendly, and then you have a maybe a more experienced kind of professional, and if you kind of tag those two people up, they they end up finding some really interesting solutions and and using technology, and you know sometimes it's technology that wasn't built for that purpose, but they found you know right where they found a way of really solving a problem by you know looking at the what's out there in terms of the possibility of technology, and what's someone that has a lot of kind of subject matter expertise, and putting those two together, and you know, sometimes that's vendor supplier kind of relationships, but often it's it's not. It's like internal researchers working together to to kind of find those those things. What are the real bottlenecks? What are the world pains that are out there. So I think construction. You know, back in early 2000s we used to go off and you know sell to and market to projects, and so people, the project director would decide whether he was going to roll out a piece of software that wasn't sustainable, and you'd have you know these toolsets with hundreds of 1000s of point solutions, and yeah, and so people got control of it, and they kind of started the innovation teams, and I think a lot of that is really positive. 00:12:50.769 --> 00:13:05.070 But I still think it's kind of you know getting out to the people in the front line, and like for you, if it's the the finance department, or for me, if it's VDC or somebody like that, then we kind of got to get out there and find out what the real bottlenecks are. 00:13:05.850 --> 00:13:14.250 Totally. Well, and I think that's like extremely tactical, and even just for our audience, I mean, we've got contractors doing 5 million in top line. 00:13:11.159 --> 00:14:12.299 We've got we've got people that are doing you know half a billion, right? So it's like I but I think the two champions is like a really tactical piece of advice, especially for like maybe a business owner that's listening to this, or someone who's wanting to scale, and they're like, "We've tried the solutions, we've looked at them, they've never really stuck. Is like lean into hiring the young people, which we'll get into a little bit more. How do we how do we attract young people right from outside of industries? It's like lean into being like we are a tech forward construction company, right? But then the second

piece is: is don't like the young person, maybe not even with like construction background or like a ton of experience. Have them come in through the fresh lens, evaluating the processes and looking how people are actually executing, and then pair that up with an experienced person who knows the inside and out why people are doing things the way that they are, to then start saying here are the highest yielding areas that we could start applying AI into our business to keep us more lean, to then potentially prevent or fill the gap of the half a million people that we need to bring into the industry. 00:14:14.070 --> 00:14:37.940 I think that's right. I mean, also just giving tools like you know, ChatGPT, Claude, like you know, I know credits and tokens and all that can get expensive, but I think you have to give those tools out to to the enterprise and let them play around those kinds of things to to find the where they you know they might start with one little thing and not wipe bubble up to being a really good kind of solution across the whole enterprise. 00:14:38.990 --> 00:14:53.980 Totally, totally. Well, and I think even from that standpoint of like again going back into like we have this like severe labor shortage in the industry, like I look at a ChatGPT or Claude at this point being more or less table stakes, right? 00:14:54.009 --> 00:15:29.629 Like if your if your company's not leveraging these types of tools, like you're just organically going to be. Behind, but even just some more tactical pieces, and again, you might not have this. Maybe I'm curious if you do. But like, what have you seen construction companies do in terms of giving their teams the autonomy that have been really successful? Is it a budget? Is it like a path to go down? Like, how do you see these companies really directing their team to say go out and find something, and we're going to give you X amount of money, or we want to pilot X number of tools, or what? What would you say that is? 00:15:30.019 --> 00:15:55.330 Yeah, I mean, I think we talked a little bit about it. You know, we've have a lot of the big kind of GCs. You know, your ENR top 400 upset up innovation teams, or digital experts, or chief digital officer, and I think they're going out and trying to understand the requirements of their teams, and then going out to the market to find reasonable solutions. And I think they've kind of come up with, you know, that's a pretty good process. 00:15:57.070 --> 00:16:04.049 But I, I'm really interested in, and I saw it a little bit in the early days, where we had kind of document controllers, right? 00:16:04.139 --> 00:16:38.000 They were the clerks and managing the drawings, and then they became kind of project information managers, PIMS. So they were kind of got into the tool sets and the tools and information more than just. So we saw kind of an evolution there, and you know, I know like Suffolk in Boston is you know putting AI engineers onto projects, and maybe

we have to find new jobs and new roles on projects to kind of help automate those processes. And maybe that's one way of being on the front line and identifying those kind of bottlenecks is you have the experts on site on projects. 00:16:38.929 --> 00:16:55.059 Totally, we see that all the time is like we could like we could sell into like a chief information officer, right? But the odds of them knowing a financial and accounting workflow to the level that they need to really validate if this is going to be a relevant solution or not is challenging. 00:16:55.690 --> 00:18:28.819 So like, is this as simple? Call it like a $50 million $100 million a year contractor. Is this as simple as like going out and finding you know a young person that's extremely tech forward with a background in whatever their operations call it to come in and then just say you're going to be boots on the ground with these guys evaluating asking questions doing these things for 612 months until you understand the problems and the process intimately or what do you see? I think so. I mean, I think that's been done a lot. Like some of the most successful implementations I've been part of, where maybe they didn't have that defined role, but there was kind of this young change agent who you know also had money boots and who kind of you know was a little ambitious, wanted to bring something new to the company, and they were often the kind of first people that we would engage in, and then they would go back to their project teams and help kind of roll it out. And you know, maybe it wasn't defined as an AI engineer in the field, but those are the guys that kind of really led the charge. And I always tried to kind of find those people, those respected kind of change agents inside of, of you know the the the contractors that I worked with, totally okay cool well and and even just a I mean I think we've covered a lot of these topics but just kind of a little bit of a sidestep and this is where your bias could come through which I'm all for but even over your 25 years of experience right working in the construction tech industry for that duration what do you think is the most important workflow for contractors to look at to automate, or maybe the highest yielding workflow for them to automate? 00:18:30.890 --> 00:18:33.829 I I think it's a big question. It's a big question. 00:18:34.069 --> 00:19:46.690 Yeah, I mean, there's so many different mission critical kind of workflows on a project, but but I think a lot of it is like lessons learned. Like people don't you know they do lots of projects. They get a lot of energy to kind of kick off a project, but they have less around the handover. Looking at the lessons learned, like what let's get really into our data and what can we learn from our data for the next project? I think a lot of that is still kind of some of the administrative kind of processes that are still, you know, a lot of multi touches and things like that. But I think we really need to be a data driven kind of industry, and and what data we have, what kind of decisions can we make from that data? You know, can we use agents? Can we use you know anything to help us kind of make better kind of project decisions? Either that's the current data set we have or the historical data set that we have. And how can we be more efficient? How can we learn from our past projects? I don't think we've really done that. You know, a lot of some people are kind of talking about that, but I don't think we've really done that on a holistic kind of perspective for the industry. 00:19:47.710 --> 00:19:55.420 Totally. Why not? Like, where where do you see the biggest bottleneck there, or what's prevented us from doing it? Is it just pure technology, or what do you think? 00:19:56.559 --> 00:20:09.119 I think there's no. I don't think it's purely technology. I think. It's the project nature of construction that you know we're up we're on to the next project. We got to kick this project off. We don't want delays. We don't want you know. 00:20:09.480 --> 00:20:33.980 So I think there's a nature of the urgency of a project, and then I don't think I think we've had kind of really disconnected data. Like you know we have data from accounting systems, we have BIM data, we have project data, we have safety data. We've kind of had these disconnected silos of point solutions, and and so we need to find ways of kind of you know pulling all that data together to to make good decisions. 00:20:34.849 --> 00:20:58.269 Totally, totally. Well, and I mean we we hear that all the time, but I mean you hear these products. We'll just use Procore as the example, right, like the big publicly traded company, right? It's like it's supposed to be the all-in-one product, right? Like the idea behind that, and a lot of consumers buy that product because they can see the vision, right? Like they experience and feel the disjointedness, but then they get into these big solutions. 00:20:54.730 --> 00:21:08.190 Like, do you think that the point solution is totally necessary, considering the nuance and variability in construction, or like, do you think there is room to build a legitimately satisfying all-in-one product? 00:21:09.690 --> 00:21:36.230 You know, it's such a hard question. That's why we're talking about it. And people have been talking about you know the point solutions versus you know best to breed versus the platform for a long time. My my take is that I think you need some fundamental foundation tools, and I I would say Procore is one of those kind of foundation kind of tools or you know competitors to Procore. 00:21:33.769 --> 00:21:56.980 Yeah, and then I think it's getting easier with AI and with integration. So you know before it was really disconnected. Now we kind of can pull that data together. We can have APIs. So I think it's a combination of your really great point solutions with a fewer few kind of fundamental kind of tools. I'm not sure that

that's the answer, but that's the best that I've come up with. 00:21:57.789 --> 00:22:14.700 Totally. Yeah. I mean, I I'm we're just saying like the the the need for a point solution, or or we could even call it like the product that's going to be best in breed, right? Like they're just like totally focused on one workflow or one component of the business, one line of business. There's still like a massive need for that. 00:22:11.369 --> 00:22:44.440 But the other side of this, and I'd be curious your take on, is like in construction, there's just the inevitable context or data that happens on the day to day that just gets missed, and the only way to track it is to get someone into a piece of software to update it in a meaningful way, right? So, what's your take on that? Like the conversations that happen the side or the superintendent, project manager that's there on a Sunday, just you know getting stuff done that they didn't during the week, and then going out on the job project and evaluating and catching things. 00:22:44.829 --> 00:22:48.490 How do you see technology being able to augment and bring that data together? 00:22:50.019 --> 00:23:00.130 Yeah, I think a lot of it, like we talked about, is kind of unstructured data. So how do you, you know, some of that's guys having a conversation, like that's data, that's information. 00:23:00.670 --> 00:23:16.680 Exactly, you know, that there's probably great, great insights there. Great information. We don't really capture that. I mean, I guess there's you know we can look at ways of using voice and and capturing meeting notes and some of those things to to capture some of that. 00:23:13.019 --> 00:23:32.539 Yeah. But I think we you know and I think the bigger companies are investing in you know one of the things that Spectral does well is it partners with a lot of the kind of big data lake kind of solutions, and we take that kind of data and bring it into the data lake and you know Databricks and things like that. 00:23:30.650 --> 00:23:53.230 So I think companies are starting to look at how they can kind of consolidate all these different you know data types into one source, and then be able to kind of bring insights and and ask questions. Be able to you know you know get insights and ask questions on that data about past projects, but also kind of predictive around what's going to happen on my my current or my future projects. 00:23:54.039 --> 00:24:08.250 Totally, totally. Well, and talk a little bit. And this is where you can like shamelessly discuss speckle because I'm curious. I hadn't heard of speckle yet, but like, do you have like what is what you talk BIM data? Give us like the high level. Like, what is it doing, and what are you the most excited about? Where do you see it driving the most value? 00:24:09.000 --> 00:24:22.160 I just you know, it's really exciting is how some of their their big customers are are using Speckle, and you know, within kind of BIM, like you have different flavors of BIM, and you get more. What are the different flavors? I'm not a big BIM guy. I should be. 00:24:22.400 --> 00:24:25.609 No, I mean like you know you have different design tools. 00:24:25.759 --> 00:25:13.470 You'll have yeah you know Revit or you'll have Navisworks or you know all these kind of different Bentley systems and you'll have all these kind of different kind of BIM tools and you'll get BIM models from different subcontractors and different designers and so how do you kind of you know QA and QC that data, but then look at that? How can you take that across all your portfolio, and how can you really get good insights out of all that BIM data? And you know what what equipment do we need for the next? You know, we talked to a retail a large really. Retailer yesterday, and they're designing you know 600 outlets. And what equipment do we need to order now so that it can be delivered? Or what formats are working best? And and and you know you could start to ask kind of your BIM data. 00:25:13.769 --> 00:25:17.069 And I think there's a lot of intelligence sitting in BIM that is just unused. 00:25:17.880 --> 00:25:41.119 Totally okay. Well, and so even just like bring this back and like it might not speckle might not like solve some of the labor shortage, but even just how you're explaining it, I hear like more like administrative office labor type work becoming much more efficient. But where do you see a product like Speckle helping impact the need for this many you know half a million new people to join the industry, or do or does it? 00:25:41.240 --> 00:26:55.210 Yeah, I think it does. I mean, we need more VDC managers. We need to bring design and construction together, you know, tighter. I mean, there's you know a lot more design build projects than there were in the past, and you know, one of the big trends I see is technology getting earlier in the cycles, and construction companies getting earlier in the cycles, and we're seeing a lot of investment in kind of pre-construction, and BIM and and you know SPUC will play in that world as well. So you know SPUCL could play across the whole kind of project lifecycle, even the digital twin and handover and all those kinds of things. So I see a big trend though where people are trying to solve problems by using this data, trying to solve problems before they become bigger problems out in the field. So I think that's one thing that's really interesting about AI and predictive analytics and and using the data efficiently is can we kind of find problems before we start you know getting out the shovels and the bulldozers and things like that, totally. And that's a that's a hell of a point. I love that point because now when you're looking at like, okay, we need half a million jobs, right? But that's half a million jobs, effectively based on how we've historically done things, right? 00:26:55.630 --> 00:27:32.150 Like we need half a million people to do the work as per defined what it was

before. Now, when we look at solutions like a speckle or an adaptive or insert whatever AI native solution, right? Because we're forward thinking, we might not need those bodies because historically, so many of those bodies were picking up the slack from poor planning, from missed data, from disjointed data, and now they're having to fill this void, so that actually would be an interesting one. You've heard it here first on builders, budgets, and beers. Do we even need the half a million, or is it closer to 300,000 Or like, what does that actually do to impact an industry? 00:27:32.809 --> 00:27:56.680 Yeah, and you know, do do we do we need a different type of person going forward? Yeah, we need different skill sets. Do we need to change our education so we get people that are ready to be in a tech-enabled kind of construction company? Do the the trade unions have to change? Do you know? Is there a merge of trade school and university? 00:27:54.789 --> 00:27:57.460 Like you know, totally. 00:27:57.940 --> 00:28:20.910 They don't have to be. I they're still kind of mutual exclusive, and you know, I I see where there could be hybrids of you know, you take some construction management courses, and you take some you know trade skill courses, and and that kind of gives you prepared for your career out in the construction industry. So I think I think that's a bigger thing that the whole industry needs to kind of think through, and you know. 00:28:20.910 --> 00:28:44.799 education, development of people, short. You know, to to help solve that kind of shortage of problem. But I think that's a bigger, like I think there's a lot of lip service to that. But I don't know if we're actually doing a lot to kind of think about what the new construction professional look like in the future, and how do we find more of those people? 00:28:45.670 --> 00:28:48.789 Totally, and that like that. 00:28:45.670 --> 00:29:42.309 This was great. I mean, because this just this kind of brought up full circle for me. Because again, I think when when you're talking to contractors, and they're they're like, "Hey, AI'S coming. We know we need to adopt it, but we're like trying to figure out how to meaningfully apply it to our business, and then when they're thinking like big picture, they're like,"Okay, great, I can make my office team as efficient as humanly possible, but I still need electricians, right? I still need people to do the work. I think this is a really good way to like understand and explain how making a tighter office team, whether that's on the BIM side or again, the accounting side to get the data that you need to make better decisions to prevent people from doing the work that they really don't like to do. I mean, and that goes back into like even the guys in the field is like how many subcontractors are frustrated when they show up to a job and what they need isn't there, what the trade before them wasn't actually complete. 00:29:39.650 --> 00:29:44.980 Now they're at a hiatus. Now that's pushing schedules, right? 00:29:42.430 --> 00:30:27.920 That's getting heartburn on the project managers, right? And then the estimators are like, we estimated this, but we're way off. And it's like this whole big issue of people doing things that they don't really want to do, but that's creating a 60-hour work week for them or whatever it is, right? That's where you can start to think about. AI on a little bit larger horizon, or even robotics for that matter, right? Of like, okay, great. Like, if we can tighten up these operations and we can get better clarity into our data, we can prevent these types of hiccups that we're traditionally seeing, and that's going to prevent us from needing to hire this many people. We can make our team more efficient. We can look at different roles that we might need, that's going to impact the business on a different level. Excuse me. So, yeah, that was super meaningful. 00:30:25.009 --> 00:30:34.460 I mean, Dax, before we wrap up here, I mean, is there any like closing thought or summary that you have for the Fine Contractors of America? 00:30:35.420 --> 00:30:51.009 No, I mean, I think it was a really you know interesting conversation, and and I really loved being part of it, but I also think it's you know we talked about some of the negative stuff. I think it's really exciting. Like we have a a chance to redesign the industry that builds America. 00:30:51.130 --> 00:31:07.289 Like we can have a different lens on how we kind of build things, and you know we can build better, faster, safer ways. And I think it's really exciting, and you know I've seen it across the last 25 years, and I'm even more excited about what could happen across the next 25 years. 00:31:08.099 --> 00:31:33.980 Totally, yeah, and I think in turn, like yeah, good good call out on like from the negative side of things, but I do I think you know even with your mention of like how does education and like the future of this generation look like is like we're we're on the cusp of, to your point, redesigning what this entire process looks like and how AI is going to fit into it. But I think, with that said, Dex, I really appreciate you jumping on, dude, and I love it. 00:31:30.200 --> 00:31:36.650 So, from here, where do if people want to follow you along, if they want to learn more from me, how do they find you? 00:31:37.279 --> 00:32:01.650 Yeah, I think LinkedIn is probably the best place to find me, and Dexter Bacheler, and I'd love-I mean, I've made so many close friendships with the companies I've worked with, my clients. You know, construction's global, so I've been on project sites in Latin America, the Middle East, China, and and that's another great part of our industry is that just the ability to make relationships and on a on a kind of global basis. So I'd love to talk to anybody. 00:32:02.279 --> 00:32:04.859 Beautiful, yeah, definitely, definitely connect with Dex. 00:32:05.339 --> 00:32:19.019 He's got he's got a lot of a lot of really good a lot of good content on LinkedIn. But cool, well, I appreciate you, sir. And yeah, we'll we'll have to make coffee or beers happen sometime since we're both in Boston. I love it. I love it. Cool, Dex. 00:32:20.099 --> 00:32:21.289 Yep, you too. Thanks for jumping on.