AJ Waters: every time we put the tech first. Things go wrong. You've got to start with the people. You've got to start by getting your end users engaged, involved, understanding how this change will affect them, we just digitized a paper form, right? And it's actually harder. To scroll through a drop down list and type on a tablet than it was to write, you know, on the piece of paper. My tech does this, AI that. Whatever. And it it doesn't really matter, right? Everybody's got features. everybody's features sound the same. Let's talk about outcomes. I love about construction is the people in this industry have one of the highest BS meters in the world. Like they know if you're blowing smoke, they know if you don't know what you're talking about. They they can see right through you. and that's good for them. It's not so good for you if you don't know what you're talking about. this country is a first-world country because of construction, right? Clean water.
Seth: Absolutely.
AJ Waters: electricity, all the things that the roads, all of that is available to us thanks to construction. I don't think that AI is gonna make it so we go out and we fire half our staffs. That I think that's that's a a a silly way to look at it. I think we'll be able to do more with the same amount of people, right? I cannot hear you.
Seth: That is why I was muted. That's such
AJ Waters: There it goes.
Seth: a typical every Teams meeting Zoom call thing, huh? I just did it.
AJ Waters: Yes.
Seth: how you doing? Happy Friday.
AJ Waters: Likewise, how are ya?
Seth: Good to see you on a screen and it's been good running into you a bunch in the last year or two, but been an avid follower as well. A lot to discuss in your personal journey. Do you mind if we could like go back? I know you've told me a lot of your background before, but I kind of want to get a version, the longest one you're willing to share if you want, of how you got where you are now. I'll pepper in some questions if You like interruptions, but start from the beginning. Like I you like Legos. We have this in common. It's the thing I say every time almost in my story is I got into this thinking I wanted to play the adult version of Legos, right? Didn't a lot of us?
AJ Waters: Yes, exactly. That was the thing. And I got into this I wanted to build stadiums. So like that was my thing. Even through high school, I was sketching stadiums and and thought that was what I was gonna do. I was gonna design and build stadiums. I did get the opportunity to do that at one point, which is so fun. It's still my favorite project, the College World Series Stadium in Omaha. But but yeah, d Legos is what I grew up with, and I was like, I wanna build things, I wanna design things, I wanna continue to play with my hands, and that's why I still build Legos today, right? It gives me that that little fix now that I've transitioned into tech of of still getting to build and and put that creativity to the test. But to answer your question, my background. Yeah, I can be really long with it or I can I can keep it pretty short. So I'll try to
Seth: We'll start with just like how did you turn up enjoyment for Legos into I should study that and make it a career? What like do you remember that thought?
AJ Waters: Yeah, great question. So I was I was going to school. I was talking about being an engineer. I was told I wasn't smart enough. so my backup plan was to go to school to be a coach. I wanted to coach basketball. And there's usually two ways to do that. Number one, you are good at it. I wasn't that good. Number two, you teach. So I went to teach math. I was good at math in high school. And so I was going to school to teach math. and lucky enough I got to take an aptitude test after my freshman year. And that aptitude test said, Hey, you'd be a good engineer. I was like, Well, that's what I wanted to do to begin with. Meanwhile, my high school sweetheart was graduating a year behind me, and she's like, I want to be a teacher. Well, I'm good at math, and teacher plus teacher doesn't equal a whole lot. So I changed my major to engineering, and that's that's how I kind of turned the love of Lego and turned it into engineering. I did Two internships at Keywit. The first year was design internship. The second year was more of a construction management internship. I I was estimating and and I found out that I liked the construction management side of it more. So even though I have a structural engineering degree, I never did get that stamp. I instead went into the field and decided to to build versus versus design. And I think
Seth: But it's cool because I'm realizing I think, and correct me if I'm wrong, but you did kind of get exposure. I mean, engineering design world, and then QIT, I would think of as kind of a builder. And then I think you've also been on the owner side and definitely into tech throughout, but now pretty dedicated to how tech can help us, right? But interesting. I don't know at which phases of your career you were in any of those roles. Do you want to start with like, like, give me the rough outline how how those things changed and what it was like starting out in the career, did you know instantly like where you should head or was it like me a wandering path, you know?
AJ Waters: Yeah. Well, it was it was a little bit of wandering and a little bit of that sounds cool. Let me try that. so I started out I came out of school full time as an estimator at Keywit in their vertical build space. So that's where I got to to learn construction and and and get out into the field some. We did time studies, we did some project control studies, but we did a lot of estimates. And that's where I got to work on the ballpark. I did that for a number of years, and actually on day two of my full-time job, I got handed a license for on-screen takeoff. I was the first person to to have a license. They said, Hey, test this. That was 07. Yep.
Seth: What year do you think this was? Sorry to interrupt. Okay, yeah, early. Yeah.
AJ Waters: And they're like, test this out. Let us know if it's any good. And then if it is, we'll roll it out to everybody else and you can teach them. So, like I led my first digital transformation, if you will, like my second day on the job. I mean, I looked at the digitizer that I had next to me and I played with on screen takeoff and it was way better. Right. And so that was kind of the first jump into it. So I I took over on screen takeoff first. I took over our our bid solution solicitation platform second. That was maybe a year or two later. So when Kewood announced that they were going to overhaul their entire IT system, I volunteered for that project. So that was one of those things where it was like, that sounds cool. The iPhone was relatively new, mobile apps were becoming a thing. I want to be a part of that. That sounds like it'll be fun. I actually was turned down. they sent someone that was more experienced, and thankfully, they needed him for a project that we won. And they had to pull him back, and I got to go later. So that that started the transition to tech. So I transit transitioned to Kiwit Technology Group in around 2012 and worked my way up there to the point where it was time for Kiwit to to spin out the stuff that they'd built internally and create in eight, right? And so when they when they moved forward with the innate project, I volunteered again and said, hey, I just spent the last three years training everybody on the stuff that we rolled out internally. I'll go help rebuild it. And I spent two years in product rebuilding the the products. And then when it went live in 2017, I transitioned into pre-sales or sales engineering, where you're the demo person, right? And and that was by sheer accident. I happened to be paying attention in all the up product update meetings that we were doing monthly. other guys would multitask or they'd only come in for their portion of the products. And I I paid attention across the whole thing. So by the time it went live, by accident I was the only guy who knew how the whole thing worked because I was listening. And and so that was one of those happy accidents where it led to a a cool new role. And then in twenty nineteen, that's where I went to work for the owner. That's where I went to Google. They called. I thought it was spam. I was like, Google can't really be calling me, right? but
Seth: Man, maybe I shouldn't ignore all those spam calls I'm getting
AJ Waters: Right? I mean, I I I did some research, I checked the person out, and they were legit. So I returned the call. they asked me to come lead their project management platform evaluation for Google Google data centers. And so I ran that project. we got to a decision, we got an implementation time frame in place. Some things didn't work out quite the way that they had been, you know, explained to me up front. So I ended up coming back to an eight. So I did boomerang around there. in 23 I I walked away from an eight and went to work for Kahua. So that that transitioned from the contractor software to the owner software. I spent a a couple of years there, really honing that that owner story in in The impact that we can have as as a as a a software in that space, probably the the last space of of construction project management that really needs to to digitize and to transform a little bit. Like the contractors have gotten there. The owners are maturing. A lot of them are are coming along behind. but there's still a lot of work to do there. And and so I was I was there and a few months ago Procor called and said, hey, we're we're gonna update the software and and innovate for owners. And sure enough they did. And I was really excited to come be a part of that that mission.
Seth: Yeah, very cool. I wonder I I kinda wanna go all the way back, but you made me think of something more recent that I just want to dig in. We'll probably sidetrack a bunch here now, but going from Kahua to Procor makes me think about like a customization versus standardization question. I'm not really sure how to put it into words, but like isn't my in Correct me if I'm wrong, but kind of my assumption is like Kahua really wanted to customize their softwares and systems pretty intentionally to each client, potentially. I'm not sure. And Procore might be on the other end of that spectrum. Am I getting it wrong? Do you and let's talk about what you're seeing there.
AJ Waters: Yeah, good question. you I think you're mostly right. Kahua is heavily configurable and their goal is for the system to be kind of custom tailored to the individual and and and that's that's a selling point for their system. ProCores is a little more out of the box, but it's a lot more configurable than people let on. Even even than I assumed before I got over here. one of the one of the misconceptions that I was taught pretty quickly once I was over on on this side of the house. And so I think there's a a level That spectrum has a fine line where you can
Seth: Mm-hmm.
AJ Waters: dance into the can customization and it makes sense for you, or you can dance into the familiarity and that makes sense for you. And and you know, no no software is gonna be perfect for everyone, but that line is. is a lot fuzzier than people realize and and you can you can overs overwhelm yourself with customization and you can oversimplify yourself with out of the box like there are pros and cons on both sides i was most excited about the the familiarity aspect of it right and the fact that when when we're talking about innovating for owners right Nine times out of 10, the people that they're inviting into their platform to collaborate with have already seen Procore, right? And they've
Seth: Mm-hmm.
AJ Waters: already used it. And that that makes that invite much easier, right? If I'm inviting you to something you've never seen before, you instantly have questions. And you've got I've I've got to get you up to speed on it. And and that that becomes a little bit more difficult. And we're talking about trying, you know, to propel the owners forward, not necessarily bog them down with additional questions. So you've got to weigh kind of both sides of that. And and there's a lot to be said on on that spectrum of customization versus
Seth: Totally.
AJ Waters: out of the box.
Seth: Yeah, we'll probably come back to it a little bit, but I g I do kinda wanna go back to the beginning and think through because I bet you will have good answers for this. Like even back to QIT. But you've been trying to help the construction industry, maybe specifically teams or or companies at times, but throughout your career, the construction industry helped them adopt tech that that actually works and provides value and leads to faster, cheaper, better. And you and I both know I feel like I've been on that journey pretty much all twenty years of my career too and and it's been as rewarding as it is frustrating. And I just wonder
AJ Waters: huh.
Seth: like what what themes do you feel like twenty years ago we were running into problems that we solved and found new problems and then solved those and found new problems or are some of them just like we can't do this? Like talk about maybe that whole spectrum like some repeating problems you saw a lot throughout your career or the hurdles like the the we can't adopt because or it doesn't work for us because what are some of the common themes you saw?
AJ Waters: Yeah, so the the common theme that that I I see more than anything else is the tech is the scapegoat, but it's never the real reason that adoption failed or that the transformation failed. what I see more often in than not, and and I've done it wrong plenty of times myself, which is how I was able to to learn this lesson, is that every time we put the tech first. Things go wrong. You've got to start with the people. You've got to start by getting your end users engaged, involved, understanding how this change will affect them, understanding what they have to do in the system, right? Any new tech can look fancy in the boardroom. Any great report, any great dashboard, right? It can look really, really good to executive leaders, but they're not pushing buttons.
Seth: Mm-hmm.
AJ Waters: And If you're going to go push that tech out, because that dashboard looked great to an executive leader, but the button clicks are twice as many as what the end user was doing before, you actually lost efficiency. You didn't gain anything, right? And so that that's something that you have to really be cognizant of and up front, understanding the people side of things and what it's going to do. To to the end user in the chair. And the input method with a lot of con tech, I think is still somewhat broken, right? That's when when I'm out at at conferences running into guys like you and and having conversations with, you know, people using these softwares every day, the thing that I still keep hearing is that we just digitized a paper form, right? And it's actually harder. To scroll through a drop down list and type on a tablet than it was to write, you know, on the piece of paper. So it it takes a little bit more time. And that's where I think a lot of the a lot of the photo AI capture ideas, if if we can really hone in on those, if we can get those right, then we start
Seth: Mm-hmm.
AJ Waters: to solve the input method problem finally.
Seth: Yeah. Real real time tracking. So like progress track that's something we've been talking about for so long. It doesn't exist. I mean, we might be getting close, I think, but
AJ Waters: Yeah.
Seth: like we talk about it as if we're doing it. We're tracking the f we know what's happening and it just so often is not even close to the case.
AJ Waters: I mean it's it's real time entered into the database, but it it's harder to do than just writing it on the piece of paper was, which obviously was delayed in h in getting back to the database, but but it's it's slowing the field down versus speeding them up, right?
Seth: Yeah. I had a a moment that reminded me of what you were just saying. in a real project scenario, I think it was only a couple weeks ago and it was an in-person OAC meeting that we were discussing like a something that was a big hurdle for the project and it was signatures on a docu sign. And everybody that needed to docusign it was actually sitting in the room physically together and they were having a technical problem. about it and they were like, guys, if this was 20 years ago, we'd be done. Everybody sign it, we'd be done. You know, like why is DocuSign causing problems for us? And I'm not trying to, I think it pr might have been somewhat user error too. So I'm not trying to badmouth DocuSign, but this happens with a lot of our tech where we haven't really thought through is this made it easier than the manual was
AJ Waters: Bingo.
Seth: Yeah. Tell me some other things that maybe you think we've solved. Like did we did we f fix anything with tech yet?
AJ Waters: Yeah, I mean, I again going back to we the real time, right? We're getting pretty close to real time, right? We're not waiting two weeks for a cost report to show up that's out of date. so so we know our productions, we know kind of where we're at, we've got a better idea. So I I think that's been solved. I think we're getting better metrics, right? In in the The companies who have who have truly looked at it from a people process tech standpoint in in in used the tech to enable the process and the people are are seeing at least more efficient, if not more collaborative workflows, right? Like we're out of email,
Seth: Hmm. Yeah.
AJ Waters: we're out of spreadsheets, we're out of this the stuff that's sitting on your hard drive, right? So if Seth won the lottery tomorrow and didn't come in to work, we're not trying to figure out where everything's at. Like those problems have all been pretty well solved. But there's I think we did create some new ones. the one that I that I talk about a lot is the the the data wars is is something that we've created because we You know, we've developed these platforms and then whoever is the host platform for the project, whether it be the contractor or the owner, one of them has the best data and the other one gets second fiddle, right? And and so I think that's another problem right now that we are engaging in actively and trying to solve by connecting the platforms in a way that we should have done from the get-go, right? but at least we figured it out now and we've we've got data transferring from one instance to the other. But yeah, those those are some of the things that we have solved, but those solves tend to create other problems.
Seth: It's something I started to think about when I was writing about trying to fix OAC meetings. I wrote a w white paper about it and I got to this point where I was thinking about a future where every stakeholder in the OAC meeting now has its own personal AI agent that advocates for their position. And if there isn't like a a project AI that actually helps the project and the community associated with the project, like you're gonna worsen. the the conflict problem in construction, because we're gonna have smarter agents fighting against other smarter agents. And I just think it compounds the the already like OAC meetings can get compati combative. And I wonder, yeah, how you create kind of like and and that's why I think about things from the owner perspective. Actually I am kind of going to midstream pivot this question because I've been thinking about it a lot. Like Which way do you build tech? From the bottom up or up from the t top down? I think like owner at the top, laborer at the bottom. And I think you have to approach it from both. I wonder if you have an opinion. I just think the owner's the ones paying the bill. And I bet you'd agree with me, nothing's really gonna get better, change, innovate unless the owner wants it to. at least probably. But at the same time, every tech you you make, it's you've said it already on this podcast. It it has to benefit the fields and the guy's actually pushing the buttons. Maybe that's a guy on a computer, but I say field, you know what I mean. So which way? Top down, bottom up, both?
AJ Waters: Well, I'd I like your both answer. one of the things or one of the stats that I like to go back to is ninety-six percent of all data collected on a construction project goes unused. That means that we developed for the bottom and they're capturing it all the way they want to, but it's useless to the top, right? now if to your point, if we go and we develop for the top and say, hey, this looks great, this is the data we want, and but capturing that data that way is cumbersome or extra clicks or super time consuming, then the field's not gonna want to do it. And even though the owner is saying contractually you're required to do it, they're gonna be upset and they're gonna give you the bare minimum that's That they can get by with so that they can move on because they're paid for time on tools, right? And so it it definitely does have to be a both ends, which historically we've as a as a con tech industry, you've either seen contractor solutions or you've seen owner solutions, right? They're not the same thing, which means they're not building on both ends. And so that's why I think it's it's important that these innovations are are coming the way they are right now, where those solutions are realizing, we've got to play both sides of this coin. because if we don't, we're hurting ourselves and our customers at one end or or the other.
Seth: Hmm. That's a good point. I was just thinking about rollouts like tech rollouts again. feels like there's a a hidden cost of doing nothing for sure. If if you're
AJ Waters: yeah.
Seth: sitting here right now as a contractor, engineer, architect, owner and you're like, yeah, AI is going crazy and I'll just wait it out, you know, see what happens and We're doing okay for now. We can wait a couple years and see what happens with A. There's definitely a cost to having that approach, but there's also a huge cost in a bad rollout. how do you get it right?
AJ Waters: Well again I I go back to starting with the people. you get it you get it right by by Knowing what you want to accomplish and running the operation in the right order, starting with the people, then going to the process, then hitting the tech. yeah, bad rollouts are painful, they're gonna happen. I think one of the issues that we have with tech implementation is is we're afraid of those bad rollouts. It's it's the old adage: nobody was ever fired for choosing IBM, right? And but but if you make the wrong tech, like if you if you make the wrong tech decision, it could be your job on the line. And we need to get past that hump and get to the point where we're okay to test and fail, right? Like not everything's gonna be perfect for us. We've gotta give it a run. And if it if it truly fails, then let's let's take a step back. But First and foremost, we need to be honest with ourselves. Was it us that failed or was it the tech that failed? And then, number two, if it was us, well, let's do it right the second time. Like let's not do that again. Let's make sure that we pivot and do it right the second time. And those are some of the lessons that we learned the hard way early on with some of our tech rollouts is we had to we had to pause and kind of start over because we we started with the tech versus starting with. what it needed to look like at the end for the people pushing the buttons. And and so we made those adjustments. But it it took longer. Yeah. It took more money. Yeah. But in the end, it worked. Right. And and so I was proud of the fact that we didn't quit. We didn't just say, this is dumb. Let's go back to the old way. We we We paused, we reset, we moved forward, right? And and there were a lot of lessons learned in doing that. And I think as an industry, we're getting more open to that idea that yeah, we're we're gonna have misses and that's gonna be okay. Because for there to really be a culture of innovation, you have to have the psychological safety to fail.
Seth: Mm-hmm. It's so true. I mean, for some reason I'm thinking about just like missionally, and this is probably why I think you and I connect the most, is like doesn't really matter to me, and I don't think to you either what company we're at or what project we're working on. Like I just think construction is so freaking cool. I love Seeing buildings
AJ Waters: yeah.
Seth: get built. And I'm also just like sad that it's filled with conflict and also has kind of a I think it's getting better. Like as we see data center projects needing labor so bad that people are like, should I become a mechanical contractor, electric? I have all these friends in the Bay Area now that are like, maybe construction is a and they've come from tech, and I'm like, Yeah, it's a fad now, but ten years ago it's funny. but Yeah, we have kind of a bad reputation as like I just think about the home contractor, the home renovating contractor. And I think it's true of anybody who says they work in construction. I think a lot of people just have a negative connotation, like they're a different class of people or something, sadly. And I think that it's wrong and off like how do you like just in general, you and I become a part of this rising tide. I think. A lot of it does there's just too much competition and hard feelings in this industry where it's like people are loyal to companies, but they're also not. And companies are it's a different time than my dad too. He worked at one engineering firm for 40 years and became a partner. And maybe I just didn't have that opportunity. But do you have advice? Like, I think there's a lot of us out there that are like, I was gonna say too, by the way. You made me think of like nothing in life comes with nothing good in life comes without pain and struggle. I think the best things come
AJ Waters: Right.
Seth: with pain and struggle. But yeah, what what advice do you have for people like me and you that are like kind of hitting our heads against a brick wall, just trying to be helpful in this industry? You know, like sure, I wanna make money, but I also genuinely want projects to go better.
AJ Waters: Yeah, that's a that's a good question. And my advice is to don't grow weary and well doing, right? for in due time you'll reap the harvest. We'll get there. And it's yeah, I mean, you know me. I sign off every blog and every podcast with construction's cool, tell your friends.
Seth: Yeah.
AJ Waters: that's that's the way that I want to go about it. And it is, it's it's a cool industry, it's an industry with purpose. It's one of the only places where you can walk around a city and say, I built that, right? And and do so with pride. And that's what I that's what I love about construction. And yes, historically, it's been a tough, rough, hard-nosed industry that that Didn't really care about people, but that's all changing. And there are plenty of voices out there that are helping it change. There are are plenty of companies out there that are helping it change, right? There's there's the conglomerate of of major ENR top 100 CEOs that have started a mental health champion team, right? So we're making the turn as an industry.
Seth: Yeah.
AJ Waters: We're making the turn. we're losing the marketing game when it comes to colleges. And and colleges saying that they're the way to the best paying jobs, they're the way, you know, your your whole construction's a fallback career or a secondhand. Like, I I agree with you. It is totally not like we get to do some of the coolest stuff, and this country is a first-world country because of construction, right? Clean water.
Seth: Absolutely.
AJ Waters: electricity, all the things that the roads, all of that is available to us thanks to construction. And so it is definitely a rewarding and fulfilling place to be. And I think I think we're just going to continue to evolve and get better now that we've recognized that we need to. And it's just a matter of continuing to change that perception that we've been kind of beaten down by the the other the other options that kids have coming out of school.
Seth: Totally. I think I've heard you joke about this too, but you brought up marketing. I think specifically about marketing tech to the construction industry. And I think I'm not disparaging you. I'm not talking about you at all. You're the kind of the opposite of this. But I've seen it a lot and I've heard you joke like sometimes I think you said software marketing teams act like they're paid by the syllable. Did you say this? but it's just there's so many acronyms,
AJ Waters: that I have been known to say that.
Seth: so many big annoying words that like especially if you're selling tech to someone in the construction industry, like b use more sense if you're selling tech to the industry. Talk about this, like how do we
AJ Waters: Yeah.
Seth: really sell tech well to construction minded people?
AJ Waters: That's that's a good question. And and I have I made that joke once before. it was in regard to a specific marketing company, not not not any of the recent ones that I've worked with, but it did feel like they were paid by the syllable, and half the time I had to Google a word to try to figure out what the heck they were saying. and that that kind of tells you where my vocabulary is, and maybe that's why I relate well. But the The fact of the matter is, how do you know how do we market to construction? What I love about construction is the people in this industry have one of the highest BS meters in the world. Like they know if you're blowing smoke, they know if you don't know what you're talking about. They they can see right through you. And and that's good for them. It's not so good for you if you don't know what you're talking about. So I think that the number one thing that we have to do when we're talking about tech and we're we're marketing to the construction industry is one, we have to understand who we're talking to. We have to understand the context of the job site and what it means to pull out a phone in that particular situation or a tablet and try to type or try to take a picture. What it means to be in freezing temperatures with gloves on and have to try to figure out. How to capture detail right at that point. And so as we start to put ourselves in their shoes, then we can start to understand, okay, this is how I talk about this particular item. But that's where you've got two roads. You talk about a feature or you talk about an outcome. And way too many folks want to go out there and just spit feature, feature, feature, feature, feature, feature, feature. My tech does this, AI that. Whatever. And it it doesn't really matter, right? Everybody's got features. everybody's features sound the same. Let's talk about outcomes. Like, what are you going to do for me? What are you going to solve? What am I going to get out of using these features? Because for me as a construction professional, I'm more worried about how I pay my guys at the end of this week. I'm more worried about how I get through next week without a major change order or how I how I look at these drawings and avoid a major clash that we didn't know was there, right? And so what are the outcomes? What are the what are the deliverables of this collection of features that your system or your tool or your technology or your AI is going to offer me to avoid that? Right.
Seth: Really good. I do want to get into AI. It is something that, like lately, I just try to say those two letters less because I'm I'm like I also I don't know. Did you see the Alex Carp interview about Palantir where he talked about the need for an application layer over-the-frontier AI models and how I think that's what I think we need to do for construction too? It's like we're not throwing stuff in chat boxes and seeing what comes out. We need to create more. deterministic workflows for our industry that are fed with really good expert in the loop curated data. and that's accomplishable. And so people do have this hang up when you start talking about AI if if you can't show them how it's actually not asking Chat GPT. Do you agree?
AJ Waters: Yeah, it's I think that's one of the common misconceptions, right? When somebody throws out AI, the first thing that people think about is Chat GPT or or Claude, right? And that that's what AI means. And that's not necessarily what AI means, especially not in every instance. I mean, Google and Netflix have been using AI to give you suggested results for decades, right? It's not that AI is brand new. It's that some of the use cases for AI are brand new. So the question that you really need to step back with as a construction professional is what are my use cases for AI, right? And then does a chatbot make sense? Or does something that's going to suggest me content make sense? Right? Because Like the idea of the Netflix AI where it, based on your history and viewing and everything that you've done, it suggests to you new shows. Like based on the history of every drawing that I've looked at and everything else, it's suggesting that this is the structural plan that I need without me having to go scroll is useful. But is that your use case? Like that's a totally different kind of AI from a chatbot. And so what The problem we've run into is everybody throws AI around like everybody knows what we're talking about. And really nobody really knows what we're talking about when we say AI. And and we don't even know what we want when we say AI because we just hear it and we're like, well, it's gotta be AI enabled. And
Seth: Yeah.
AJ Waters: and so it's it's time for us all to take a step back and say, all right, what do we want again as an outcome from
Seth: Mm-hmm.
AJ Waters: the AI? What is it that we want it to do? All right, now what's the right kind of AI to deploy? Chatbot, contextual search, historical search, whatever.
Seth: Yeah, very cool. Tell me about like your latest experiences. How are you liking Procore? Anything you wanna share, like what you're specifically working on? Do you wanna talk about systems of record? I don't know. I'll kinda leave it open to you.
AJ Waters: Well, the systems of record are are something that I I do can kinda When since we're on the AI topic, right,
Seth: Yeah.
AJ Waters: I think something that people tend to forget whenever they talk about throwing just a ontology layer over the top of data is where the heck is the data, right?
Seth: Mm-hmm.
AJ Waters: Because it doesn't just float around in a cloud. That's not how it really works, right? It's gotta be contextualized and organized and in a way that it's useful. And that's where systems of record come into play, right? or systems of truth. is another way to talk about it. And so I think I think systems of record are still going to be vital in the AI age. There's gonna be the record and then the action. And the better that they work with one another, the better experience you're gonna have. Think about it. Why has Apple been so successful with the iPhone? Because the system of action. Is built on the system of record. They're not built by two different companies, right? It's it's a a
Seth: Makes sense.
AJ Waters: closed system. And yeah, there are people who are against that, but by golly, they succeeded with it. And it's very easy to use, right? It's incredibly easy to use and it and it works over the top of itself. So that's that's what I'm really excited about, is and why I think the system of record is still going to be valuable because the data has to sit somewhere. But but to answer your question about you know how am I liking Procore and what are we working on? So I'm I'm loving Procore, number one. Great place to work. it's it's been really exciting. The people here are great. I've I've met some amazing folks. I've I've I'm getting up to speed faster than I thought I would, which is nice. and a lot of that has to do with the amount of people that are willing to help, but My my focus here at Procore is on the public sector and the public sector owners, which which is a different kind of animal from the private sector, right? They they operate differently, they have different procurement models, they have much stricter budgets, like those sorts of things. So we've gotta we've gotta get creative with that. But again, on the flip side, when you're talking about Where in the industry is there the most opportunity to still make a significant impact? That's a part of the industry that I think could really be a place where we could provide better insights, better workflows, better systems of record than what they have historically been been left with. And that's that's why I'm so excited to be here and and be a part of this particular team.
Seth: It feels like there's a future pretty soon the horizon here where an owner like I like to imagine it in a a homeowner scenario. Like let's say I wanted to build my dream house somewhere in the hills in Colorado and I could just bring my phone and talk it through, I found this site, let's take a video of it. I want a house like this with my master bedroom over here and a window that lets the sun rise and set over you know, like think through the project and then it would be like, all right, I got your your bridging documents ready for you to go out and bid it to contractors and it will already send it to contractors and receive bids and and pick the the best one, the best value and and get them signed up and then like you put a camera on site and it you can know you can just ask it, is are the beams in yet? Any issues with waterproofing, you know, and it will tell you what you need to know. so I think AI is gonna handle so many things so fast that I I I want you to kind of talk me through how you if you think you could like imagine the next five years and then like how are we gonna transition into a place where there's a lot less people needing to be involved in this? Or do you think that's just fantasy?
AJ Waters: I don't here's the thing. I don't think there's less people that need to be involved. I think
Seth: Sure.
AJ Waters: their involvement changes. Right? Like I don't think that AI is gonna make it so we go out and we fire half our staffs. That I think that's that's a a a silly way to look at it. I think we'll be able to do more with the same amount of people, right? Like that's That's what I think the the goal with tech should be to make us more efficient to where we can do more with the same. We can scale the business faster. there's there's a a part of what you explained there that I like. That's setting up the camera and just being able to ask AI what's the progress of the of the build. I think we're getting close to stuff like that with some of the the the progress input methods that we talked about earlier. There's a part of it that I'm not so keen on yet. And and let me tell you why. And that's the walk up with the camera and say, here's here's the the building I want, and I want the office space over here and this over there and that, and have it create the bridging documents. There's still a missing piece in the middle there, right? Because like I can I can verbally articulate what I think I want, but until I start to see it. I won't really know, right? Because I can I can think that I'm saying something. Like, how many times
Seth: Totally.
AJ Waters: is what you said been lost in translation by what someone else heard? And e and a AI is no different. I was actually having this conversation this morning with a with a friend of mine on on our podcast. Like he he builds custom things like this. He built this built in behind me. And he was building one for a friend. They they gave him every detail that they wanted verbally. He kind of sketched it up. They they said, Yeah, that looks great. He gave him progress photos. By the time it was done and they installed it, it ended up being not what they wanted, right? Because they didn't really see it. And so that's where I think the bridge becomes we finally get to a place where the model is useful, right?
Seth: Yeah.
AJ Waters: And and that becomes kind of that bridge of what I articulated to you is what I'm now seeing. And so augmented reality that can put that in place where I'm standing on the site and I can see the building there, but it's not, right? That kind of concept will be will be a big win before we get to anything where we're just trying to explain the project out.
Seth: You're very right. And it does make me think the more I work with AI, the more I realize how important like a feedback loop or a a lessons learned
AJ Waters: Mm, mm-hmm.
Seth: type situ like you have to enforce some kind of like a okay, that's not how I like it. Show me a new draft a little bit more like this. And you have to iterate a bunch of times and even like I I sometimes start with a prompt and iterate on it to the point where I I realize 10 prompts in. I should have just prompted with this at the beginning, but it's like the system was helping me figure out a more powerful prompt as I went on. But that feedback loop I think is extremely important. Obviously, we need human judgment involved too. I do think I'm skipping a lot when I say AI will just create the design for you. Well, I still think there's a place for engineers who know the the local soil. And people
AJ Waters: Right?
Seth: who understand the regulations in those particular areas. And then even just like letting it flat out select contractors or designers or anybody else, like there's a human judgment layer in there that we're always gonna want. I talk about it a lot. You said the one neck to choke thing. I totally use that a lot, even though it is kind of negative. It's still, it's just like people, I don't think owners of big building projects are gonna be like, yeah, I'll just let AI do it. I mean,
AJ Waters: Right.
Seth: I think we're gonna have people in front of us for a long time. So I agree with you. You talk about that too, like in these change management moments that like a lot of it has to do with finding people who kind of have a superpower for it and and giving them empowerment for it. Talk about that. Like how do you find the the advocates and what are they good at? What are people in company in companies that succeed with a rollout? They have somebody who kind of like takes it under their wing.
AJ Waters: Yeah, so there's there's two kinds of champions that I think every major successful rollout needs. Number one, they need the influential, low level, uping up and comer, right? They're usually they're probably on your your succession plan somewhere or you you see them as as an up and coming leader, but they have influence among their peers, right? So when When they say stuff, people listen. not in a way that they like beat them down, but they're likable, right? They have charisma. That that that sort of you need that sort of individual. And then you need a similar version of that at the executive level. And a lot of people think that the executive sponsor is there to like beat the hammer and drive down through the organization. This change is coming and we're gonna change, and you're all on board or else, right? But it's actually the exact opposite. The executive sponsors' number one job is to get their peers on board. Because there's nothing worse than a district or area manager that says, we're not doing that. I don't care what they say at corporate. We're going to keep doing it our way. Don't worry about it. Because that creates more trouble than anything else. And so that That executive sponsor has to get their peer group on board with the change as well, explain what's in it for their peers. And and I think that often gets overlooked, like that little piece of of that executive sponsor's influence gets overlooked. Yeah, because at that level, how many times is it competitive, right? Like my district's up against yours for bonuses and and prizes and all that kind of stuff. and and I'm I'm gonna keep doing it my way because I know that I'm not gonna slow down. You're gonna have to slow down as you implement and do all this stuff. And and so that's that's vitally important that you have champions at each level.
Seth: Hmm. That's good. Well, I'm sure I'll I'll probably bug you again in six months or a year or something and try to get you to come on and do an update. So I hope you stay tuned for that. And do you wanna plug anything? Co reach out to AJ on LinkedIn. Anything else?
AJ Waters: That's right. Yeah, look me up on LinkedIn. It's AJ Waters, Super Simple. and and I'm always willing to meet new folks in in the industry.
Seth: Very cool. Thanks, Adrian.
AJ Waters: Yeah, you bet.
Seth: Absolutely.
AJ Waters: electricity, all the things that the roads, all of that is available to us thanks to construction. I don't think that AI is gonna make it so we go out and we fire half our staffs. That I think that's that's a a a silly way to look at it. I think we'll be able to do more with the same amount of people, right? I cannot hear you.
Seth: That is why I was muted. That's such
AJ Waters: There it goes.
Seth: a typical every Teams meeting Zoom call thing, huh? I just did it.
AJ Waters: Yes.
Seth: how you doing? Happy Friday.
AJ Waters: Likewise, how are ya?
Seth: Good to see you on a screen and it's been good running into you a bunch in the last year or two, but been an avid follower as well. A lot to discuss in your personal journey. Do you mind if we could like go back? I know you've told me a lot of your background before, but I kind of want to get a version, the longest one you're willing to share if you want, of how you got where you are now. I'll pepper in some questions if You like interruptions, but start from the beginning. Like I you like Legos. We have this in common. It's the thing I say every time almost in my story is I got into this thinking I wanted to play the adult version of Legos, right? Didn't a lot of us?
AJ Waters: Yes, exactly. That was the thing. And I got into this I wanted to build stadiums. So like that was my thing. Even through high school, I was sketching stadiums and and thought that was what I was gonna do. I was gonna design and build stadiums. I did get the opportunity to do that at one point, which is so fun. It's still my favorite project, the College World Series Stadium in Omaha. But but yeah, d Legos is what I grew up with, and I was like, I wanna build things, I wanna design things, I wanna continue to play with my hands, and that's why I still build Legos today, right? It gives me that that little fix now that I've transitioned into tech of of still getting to build and and put that creativity to the test. But to answer your question, my background. Yeah, I can be really long with it or I can I can keep it pretty short. So I'll try to
Seth: We'll start with just like how did you turn up enjoyment for Legos into I should study that and make it a career? What like do you remember that thought?
AJ Waters: Yeah, great question. So I was I was going to school. I was talking about being an engineer. I was told I wasn't smart enough. so my backup plan was to go to school to be a coach. I wanted to coach basketball. And there's usually two ways to do that. Number one, you are good at it. I wasn't that good. Number two, you teach. So I went to teach math. I was good at math in high school. And so I was going to school to teach math. and lucky enough I got to take an aptitude test after my freshman year. And that aptitude test said, Hey, you'd be a good engineer. I was like, Well, that's what I wanted to do to begin with. Meanwhile, my high school sweetheart was graduating a year behind me, and she's like, I want to be a teacher. Well, I'm good at math, and teacher plus teacher doesn't equal a whole lot. So I changed my major to engineering, and that's that's how I kind of turned the love of Lego and turned it into engineering. I did Two internships at Keywit. The first year was design internship. The second year was more of a construction management internship. I I was estimating and and I found out that I liked the construction management side of it more. So even though I have a structural engineering degree, I never did get that stamp. I instead went into the field and decided to to build versus versus design. And I think
Seth: But it's cool because I'm realizing I think, and correct me if I'm wrong, but you did kind of get exposure. I mean, engineering design world, and then QIT, I would think of as kind of a builder. And then I think you've also been on the owner side and definitely into tech throughout, but now pretty dedicated to how tech can help us, right? But interesting. I don't know at which phases of your career you were in any of those roles. Do you want to start with like, like, give me the rough outline how how those things changed and what it was like starting out in the career, did you know instantly like where you should head or was it like me a wandering path, you know?
AJ Waters: Yeah. Well, it was it was a little bit of wandering and a little bit of that sounds cool. Let me try that. so I started out I came out of school full time as an estimator at Keywit in their vertical build space. So that's where I got to to learn construction and and and get out into the field some. We did time studies, we did some project control studies, but we did a lot of estimates. And that's where I got to work on the ballpark. I did that for a number of years, and actually on day two of my full-time job, I got handed a license for on-screen takeoff. I was the first person to to have a license. They said, Hey, test this. That was 07. Yep.
Seth: What year do you think this was? Sorry to interrupt. Okay, yeah, early. Yeah.
AJ Waters: And they're like, test this out. Let us know if it's any good. And then if it is, we'll roll it out to everybody else and you can teach them. So, like I led my first digital transformation, if you will, like my second day on the job. I mean, I looked at the digitizer that I had next to me and I played with on screen takeoff and it was way better. Right. And so that was kind of the first jump into it. So I I took over on screen takeoff first. I took over our our bid solution solicitation platform second. That was maybe a year or two later. So when Kewood announced that they were going to overhaul their entire IT system, I volunteered for that project. So that was one of those things where it was like, that sounds cool. The iPhone was relatively new, mobile apps were becoming a thing. I want to be a part of that. That sounds like it'll be fun. I actually was turned down. they sent someone that was more experienced, and thankfully, they needed him for a project that we won. And they had to pull him back, and I got to go later. So that that started the transition to tech. So I transit transitioned to Kiwit Technology Group in around 2012 and worked my way up there to the point where it was time for Kiwit to to spin out the stuff that they'd built internally and create in eight, right? And so when they when they moved forward with the innate project, I volunteered again and said, hey, I just spent the last three years training everybody on the stuff that we rolled out internally. I'll go help rebuild it. And I spent two years in product rebuilding the the products. And then when it went live in 2017, I transitioned into pre-sales or sales engineering, where you're the demo person, right? And and that was by sheer accident. I happened to be paying attention in all the up product update meetings that we were doing monthly. other guys would multitask or they'd only come in for their portion of the products. And I I paid attention across the whole thing. So by the time it went live, by accident I was the only guy who knew how the whole thing worked because I was listening. And and so that was one of those happy accidents where it led to a a cool new role. And then in twenty nineteen, that's where I went to work for the owner. That's where I went to Google. They called. I thought it was spam. I was like, Google can't really be calling me, right? but
Seth: Man, maybe I shouldn't ignore all those spam calls I'm getting
AJ Waters: Right? I mean, I I I did some research, I checked the person out, and they were legit. So I returned the call. they asked me to come lead their project management platform evaluation for Google Google data centers. And so I ran that project. we got to a decision, we got an implementation time frame in place. Some things didn't work out quite the way that they had been, you know, explained to me up front. So I ended up coming back to an eight. So I did boomerang around there. in 23 I I walked away from an eight and went to work for Kahua. So that that transitioned from the contractor software to the owner software. I spent a a couple of years there, really honing that that owner story in in The impact that we can have as as a as a a software in that space, probably the the last space of of construction project management that really needs to to digitize and to transform a little bit. Like the contractors have gotten there. The owners are maturing. A lot of them are are coming along behind. but there's still a lot of work to do there. And and so I was I was there and a few months ago Procor called and said, hey, we're we're gonna update the software and and innovate for owners. And sure enough they did. And I was really excited to come be a part of that that mission.
Seth: Yeah, very cool. I wonder I I kinda wanna go all the way back, but you made me think of something more recent that I just want to dig in. We'll probably sidetrack a bunch here now, but going from Kahua to Procor makes me think about like a customization versus standardization question. I'm not really sure how to put it into words, but like isn't my in Correct me if I'm wrong, but kind of my assumption is like Kahua really wanted to customize their softwares and systems pretty intentionally to each client, potentially. I'm not sure. And Procore might be on the other end of that spectrum. Am I getting it wrong? Do you and let's talk about what you're seeing there.
AJ Waters: Yeah, good question. you I think you're mostly right. Kahua is heavily configurable and their goal is for the system to be kind of custom tailored to the individual and and and that's that's a selling point for their system. ProCores is a little more out of the box, but it's a lot more configurable than people let on. Even even than I assumed before I got over here. one of the one of the misconceptions that I was taught pretty quickly once I was over on on this side of the house. And so I think there's a a level That spectrum has a fine line where you can
Seth: Mm-hmm.
AJ Waters: dance into the can customization and it makes sense for you, or you can dance into the familiarity and that makes sense for you. And and you know, no no software is gonna be perfect for everyone, but that line is. is a lot fuzzier than people realize and and you can you can overs overwhelm yourself with customization and you can oversimplify yourself with out of the box like there are pros and cons on both sides i was most excited about the the familiarity aspect of it right and the fact that when when we're talking about innovating for owners right Nine times out of 10, the people that they're inviting into their platform to collaborate with have already seen Procore, right? And they've
Seth: Mm-hmm.
AJ Waters: already used it. And that that makes that invite much easier, right? If I'm inviting you to something you've never seen before, you instantly have questions. And you've got I've I've got to get you up to speed on it. And and that that becomes a little bit more difficult. And we're talking about trying, you know, to propel the owners forward, not necessarily bog them down with additional questions. So you've got to weigh kind of both sides of that. And and there's a lot to be said on on that spectrum of customization versus
Seth: Totally.
AJ Waters: out of the box.
Seth: Yeah, we'll probably come back to it a little bit, but I g I do kinda wanna go back to the beginning and think through because I bet you will have good answers for this. Like even back to QIT. But you've been trying to help the construction industry, maybe specifically teams or or companies at times, but throughout your career, the construction industry helped them adopt tech that that actually works and provides value and leads to faster, cheaper, better. And you and I both know I feel like I've been on that journey pretty much all twenty years of my career too and and it's been as rewarding as it is frustrating. And I just wonder
AJ Waters: huh.
Seth: like what what themes do you feel like twenty years ago we were running into problems that we solved and found new problems and then solved those and found new problems or are some of them just like we can't do this? Like talk about maybe that whole spectrum like some repeating problems you saw a lot throughout your career or the hurdles like the the we can't adopt because or it doesn't work for us because what are some of the common themes you saw?
AJ Waters: Yeah, so the the common theme that that I I see more than anything else is the tech is the scapegoat, but it's never the real reason that adoption failed or that the transformation failed. what I see more often in than not, and and I've done it wrong plenty of times myself, which is how I was able to to learn this lesson, is that every time we put the tech first. Things go wrong. You've got to start with the people. You've got to start by getting your end users engaged, involved, understanding how this change will affect them, understanding what they have to do in the system, right? Any new tech can look fancy in the boardroom. Any great report, any great dashboard, right? It can look really, really good to executive leaders, but they're not pushing buttons.
Seth: Mm-hmm.
AJ Waters: And If you're going to go push that tech out, because that dashboard looked great to an executive leader, but the button clicks are twice as many as what the end user was doing before, you actually lost efficiency. You didn't gain anything, right? And so that that's something that you have to really be cognizant of and up front, understanding the people side of things and what it's going to do. To to the end user in the chair. And the input method with a lot of con tech, I think is still somewhat broken, right? That's when when I'm out at at conferences running into guys like you and and having conversations with, you know, people using these softwares every day, the thing that I still keep hearing is that we just digitized a paper form, right? And it's actually harder. To scroll through a drop down list and type on a tablet than it was to write, you know, on the piece of paper. So it it takes a little bit more time. And that's where I think a lot of the a lot of the photo AI capture ideas, if if we can really hone in on those, if we can get those right, then we start
Seth: Mm-hmm.
AJ Waters: to solve the input method problem finally.
Seth: Yeah. Real real time tracking. So like progress track that's something we've been talking about for so long. It doesn't exist. I mean, we might be getting close, I think, but
AJ Waters: Yeah.
Seth: like we talk about it as if we're doing it. We're tracking the f we know what's happening and it just so often is not even close to the case.
AJ Waters: I mean it's it's real time entered into the database, but it it's harder to do than just writing it on the piece of paper was, which obviously was delayed in h in getting back to the database, but but it's it's slowing the field down versus speeding them up, right?
Seth: Yeah. I had a a moment that reminded me of what you were just saying. in a real project scenario, I think it was only a couple weeks ago and it was an in-person OAC meeting that we were discussing like a something that was a big hurdle for the project and it was signatures on a docu sign. And everybody that needed to docusign it was actually sitting in the room physically together and they were having a technical problem. about it and they were like, guys, if this was 20 years ago, we'd be done. Everybody sign it, we'd be done. You know, like why is DocuSign causing problems for us? And I'm not trying to, I think it pr might have been somewhat user error too. So I'm not trying to badmouth DocuSign, but this happens with a lot of our tech where we haven't really thought through is this made it easier than the manual was
AJ Waters: Bingo.
Seth: Yeah. Tell me some other things that maybe you think we've solved. Like did we did we f fix anything with tech yet?
AJ Waters: Yeah, I mean, I again going back to we the real time, right? We're getting pretty close to real time, right? We're not waiting two weeks for a cost report to show up that's out of date. so so we know our productions, we know kind of where we're at, we've got a better idea. So I I think that's been solved. I think we're getting better metrics, right? In in the The companies who have who have truly looked at it from a people process tech standpoint in in in used the tech to enable the process and the people are are seeing at least more efficient, if not more collaborative workflows, right? Like we're out of email,
Seth: Hmm. Yeah.
AJ Waters: we're out of spreadsheets, we're out of this the stuff that's sitting on your hard drive, right? So if Seth won the lottery tomorrow and didn't come in to work, we're not trying to figure out where everything's at. Like those problems have all been pretty well solved. But there's I think we did create some new ones. the one that I that I talk about a lot is the the the data wars is is something that we've created because we You know, we've developed these platforms and then whoever is the host platform for the project, whether it be the contractor or the owner, one of them has the best data and the other one gets second fiddle, right? And and so I think that's another problem right now that we are engaging in actively and trying to solve by connecting the platforms in a way that we should have done from the get-go, right? but at least we figured it out now and we've we've got data transferring from one instance to the other. But yeah, those those are some of the things that we have solved, but those solves tend to create other problems.
Seth: It's something I started to think about when I was writing about trying to fix OAC meetings. I wrote a w white paper about it and I got to this point where I was thinking about a future where every stakeholder in the OAC meeting now has its own personal AI agent that advocates for their position. And if there isn't like a a project AI that actually helps the project and the community associated with the project, like you're gonna worsen. the the conflict problem in construction, because we're gonna have smarter agents fighting against other smarter agents. And I just think it compounds the the already like OAC meetings can get compati combative. And I wonder, yeah, how you create kind of like and and that's why I think about things from the owner perspective. Actually I am kind of going to midstream pivot this question because I've been thinking about it a lot. Like Which way do you build tech? From the bottom up or up from the t top down? I think like owner at the top, laborer at the bottom. And I think you have to approach it from both. I wonder if you have an opinion. I just think the owner's the ones paying the bill. And I bet you'd agree with me, nothing's really gonna get better, change, innovate unless the owner wants it to. at least probably. But at the same time, every tech you you make, it's you've said it already on this podcast. It it has to benefit the fields and the guy's actually pushing the buttons. Maybe that's a guy on a computer, but I say field, you know what I mean. So which way? Top down, bottom up, both?
AJ Waters: Well, I'd I like your both answer. one of the things or one of the stats that I like to go back to is ninety-six percent of all data collected on a construction project goes unused. That means that we developed for the bottom and they're capturing it all the way they want to, but it's useless to the top, right? now if to your point, if we go and we develop for the top and say, hey, this looks great, this is the data we want, and but capturing that data that way is cumbersome or extra clicks or super time consuming, then the field's not gonna want to do it. And even though the owner is saying contractually you're required to do it, they're gonna be upset and they're gonna give you the bare minimum that's That they can get by with so that they can move on because they're paid for time on tools, right? And so it it definitely does have to be a both ends, which historically we've as a as a con tech industry, you've either seen contractor solutions or you've seen owner solutions, right? They're not the same thing, which means they're not building on both ends. And so that's why I think it's it's important that these innovations are are coming the way they are right now, where those solutions are realizing, we've got to play both sides of this coin. because if we don't, we're hurting ourselves and our customers at one end or or the other.
Seth: Hmm. That's a good point. I was just thinking about rollouts like tech rollouts again. feels like there's a a hidden cost of doing nothing for sure. If if you're
AJ Waters: yeah.
Seth: sitting here right now as a contractor, engineer, architect, owner and you're like, yeah, AI is going crazy and I'll just wait it out, you know, see what happens and We're doing okay for now. We can wait a couple years and see what happens with A. There's definitely a cost to having that approach, but there's also a huge cost in a bad rollout. how do you get it right?
AJ Waters: Well again I I go back to starting with the people. you get it you get it right by by Knowing what you want to accomplish and running the operation in the right order, starting with the people, then going to the process, then hitting the tech. yeah, bad rollouts are painful, they're gonna happen. I think one of the issues that we have with tech implementation is is we're afraid of those bad rollouts. It's it's the old adage: nobody was ever fired for choosing IBM, right? And but but if you make the wrong tech, like if you if you make the wrong tech decision, it could be your job on the line. And we need to get past that hump and get to the point where we're okay to test and fail, right? Like not everything's gonna be perfect for us. We've gotta give it a run. And if it if it truly fails, then let's let's take a step back. But First and foremost, we need to be honest with ourselves. Was it us that failed or was it the tech that failed? And then, number two, if it was us, well, let's do it right the second time. Like let's not do that again. Let's make sure that we pivot and do it right the second time. And those are some of the lessons that we learned the hard way early on with some of our tech rollouts is we had to we had to pause and kind of start over because we we started with the tech versus starting with. what it needed to look like at the end for the people pushing the buttons. And and so we made those adjustments. But it it took longer. Yeah. It took more money. Yeah. But in the end, it worked. Right. And and so I was proud of the fact that we didn't quit. We didn't just say, this is dumb. Let's go back to the old way. We we We paused, we reset, we moved forward, right? And and there were a lot of lessons learned in doing that. And I think as an industry, we're getting more open to that idea that yeah, we're we're gonna have misses and that's gonna be okay. Because for there to really be a culture of innovation, you have to have the psychological safety to fail.
Seth: Mm-hmm. It's so true. I mean, for some reason I'm thinking about just like missionally, and this is probably why I think you and I connect the most, is like doesn't really matter to me, and I don't think to you either what company we're at or what project we're working on. Like I just think construction is so freaking cool. I love Seeing buildings
AJ Waters: yeah.
Seth: get built. And I'm also just like sad that it's filled with conflict and also has kind of a I think it's getting better. Like as we see data center projects needing labor so bad that people are like, should I become a mechanical contractor, electric? I have all these friends in the Bay Area now that are like, maybe construction is a and they've come from tech, and I'm like, Yeah, it's a fad now, but ten years ago it's funny. but Yeah, we have kind of a bad reputation as like I just think about the home contractor, the home renovating contractor. And I think it's true of anybody who says they work in construction. I think a lot of people just have a negative connotation, like they're a different class of people or something, sadly. And I think that it's wrong and off like how do you like just in general, you and I become a part of this rising tide. I think. A lot of it does there's just too much competition and hard feelings in this industry where it's like people are loyal to companies, but they're also not. And companies are it's a different time than my dad too. He worked at one engineering firm for 40 years and became a partner. And maybe I just didn't have that opportunity. But do you have advice? Like, I think there's a lot of us out there that are like, I was gonna say too, by the way. You made me think of like nothing in life comes with nothing good in life comes without pain and struggle. I think the best things come
AJ Waters: Right.
Seth: with pain and struggle. But yeah, what what advice do you have for people like me and you that are like kind of hitting our heads against a brick wall, just trying to be helpful in this industry? You know, like sure, I wanna make money, but I also genuinely want projects to go better.
AJ Waters: Yeah, that's a that's a good question. And my advice is to don't grow weary and well doing, right? for in due time you'll reap the harvest. We'll get there. And it's yeah, I mean, you know me. I sign off every blog and every podcast with construction's cool, tell your friends.
Seth: Yeah.
AJ Waters: that's that's the way that I want to go about it. And it is, it's it's a cool industry, it's an industry with purpose. It's one of the only places where you can walk around a city and say, I built that, right? And and do so with pride. And that's what I that's what I love about construction. And yes, historically, it's been a tough, rough, hard-nosed industry that that Didn't really care about people, but that's all changing. And there are plenty of voices out there that are helping it change. There are are plenty of companies out there that are helping it change, right? There's there's the conglomerate of of major ENR top 100 CEOs that have started a mental health champion team, right? So we're making the turn as an industry.
Seth: Yeah.
AJ Waters: We're making the turn. we're losing the marketing game when it comes to colleges. And and colleges saying that they're the way to the best paying jobs, they're the way, you know, your your whole construction's a fallback career or a secondhand. Like, I I agree with you. It is totally not like we get to do some of the coolest stuff, and this country is a first-world country because of construction, right? Clean water.
Seth: Absolutely.
AJ Waters: electricity, all the things that the roads, all of that is available to us thanks to construction. And so it is definitely a rewarding and fulfilling place to be. And I think I think we're just going to continue to evolve and get better now that we've recognized that we need to. And it's just a matter of continuing to change that perception that we've been kind of beaten down by the the other the other options that kids have coming out of school.
Seth: Totally. I think I've heard you joke about this too, but you brought up marketing. I think specifically about marketing tech to the construction industry. And I think I'm not disparaging you. I'm not talking about you at all. You're the kind of the opposite of this. But I've seen it a lot and I've heard you joke like sometimes I think you said software marketing teams act like they're paid by the syllable. Did you say this? but it's just there's so many acronyms,
AJ Waters: that I have been known to say that.
Seth: so many big annoying words that like especially if you're selling tech to someone in the construction industry, like b use more sense if you're selling tech to the industry. Talk about this, like how do we
AJ Waters: Yeah.
Seth: really sell tech well to construction minded people?
AJ Waters: That's that's a good question. And and I have I made that joke once before. it was in regard to a specific marketing company, not not not any of the recent ones that I've worked with, but it did feel like they were paid by the syllable, and half the time I had to Google a word to try to figure out what the heck they were saying. and that that kind of tells you where my vocabulary is, and maybe that's why I relate well. But the The fact of the matter is, how do you know how do we market to construction? What I love about construction is the people in this industry have one of the highest BS meters in the world. Like they know if you're blowing smoke, they know if you don't know what you're talking about. They they can see right through you. And and that's good for them. It's not so good for you if you don't know what you're talking about. So I think that the number one thing that we have to do when we're talking about tech and we're we're marketing to the construction industry is one, we have to understand who we're talking to. We have to understand the context of the job site and what it means to pull out a phone in that particular situation or a tablet and try to type or try to take a picture. What it means to be in freezing temperatures with gloves on and have to try to figure out. How to capture detail right at that point. And so as we start to put ourselves in their shoes, then we can start to understand, okay, this is how I talk about this particular item. But that's where you've got two roads. You talk about a feature or you talk about an outcome. And way too many folks want to go out there and just spit feature, feature, feature, feature, feature, feature, feature. My tech does this, AI that. Whatever. And it it doesn't really matter, right? Everybody's got features. everybody's features sound the same. Let's talk about outcomes. Like, what are you going to do for me? What are you going to solve? What am I going to get out of using these features? Because for me as a construction professional, I'm more worried about how I pay my guys at the end of this week. I'm more worried about how I get through next week without a major change order or how I how I look at these drawings and avoid a major clash that we didn't know was there, right? And so what are the outcomes? What are the what are the deliverables of this collection of features that your system or your tool or your technology or your AI is going to offer me to avoid that? Right.
Seth: Really good. I do want to get into AI. It is something that, like lately, I just try to say those two letters less because I'm I'm like I also I don't know. Did you see the Alex Carp interview about Palantir where he talked about the need for an application layer over-the-frontier AI models and how I think that's what I think we need to do for construction too? It's like we're not throwing stuff in chat boxes and seeing what comes out. We need to create more. deterministic workflows for our industry that are fed with really good expert in the loop curated data. and that's accomplishable. And so people do have this hang up when you start talking about AI if if you can't show them how it's actually not asking Chat GPT. Do you agree?
AJ Waters: Yeah, it's I think that's one of the common misconceptions, right? When somebody throws out AI, the first thing that people think about is Chat GPT or or Claude, right? And that that's what AI means. And that's not necessarily what AI means, especially not in every instance. I mean, Google and Netflix have been using AI to give you suggested results for decades, right? It's not that AI is brand new. It's that some of the use cases for AI are brand new. So the question that you really need to step back with as a construction professional is what are my use cases for AI, right? And then does a chatbot make sense? Or does something that's going to suggest me content make sense? Right? Because Like the idea of the Netflix AI where it, based on your history and viewing and everything that you've done, it suggests to you new shows. Like based on the history of every drawing that I've looked at and everything else, it's suggesting that this is the structural plan that I need without me having to go scroll is useful. But is that your use case? Like that's a totally different kind of AI from a chatbot. And so what The problem we've run into is everybody throws AI around like everybody knows what we're talking about. And really nobody really knows what we're talking about when we say AI. And and we don't even know what we want when we say AI because we just hear it and we're like, well, it's gotta be AI enabled. And
Seth: Yeah.
AJ Waters: and so it's it's time for us all to take a step back and say, all right, what do we want again as an outcome from
Seth: Mm-hmm.
AJ Waters: the AI? What is it that we want it to do? All right, now what's the right kind of AI to deploy? Chatbot, contextual search, historical search, whatever.
Seth: Yeah, very cool. Tell me about like your latest experiences. How are you liking Procore? Anything you wanna share, like what you're specifically working on? Do you wanna talk about systems of record? I don't know. I'll kinda leave it open to you.
AJ Waters: Well, the systems of record are are something that I I do can kinda When since we're on the AI topic, right,
Seth: Yeah.
AJ Waters: I think something that people tend to forget whenever they talk about throwing just a ontology layer over the top of data is where the heck is the data, right?
Seth: Mm-hmm.
AJ Waters: Because it doesn't just float around in a cloud. That's not how it really works, right? It's gotta be contextualized and organized and in a way that it's useful. And that's where systems of record come into play, right? or systems of truth. is another way to talk about it. And so I think I think systems of record are still going to be vital in the AI age. There's gonna be the record and then the action. And the better that they work with one another, the better experience you're gonna have. Think about it. Why has Apple been so successful with the iPhone? Because the system of action. Is built on the system of record. They're not built by two different companies, right? It's it's a a
Seth: Makes sense.
AJ Waters: closed system. And yeah, there are people who are against that, but by golly, they succeeded with it. And it's very easy to use, right? It's incredibly easy to use and it and it works over the top of itself. So that's that's what I'm really excited about, is and why I think the system of record is still going to be valuable because the data has to sit somewhere. But but to answer your question about you know how am I liking Procore and what are we working on? So I'm I'm loving Procore, number one. Great place to work. it's it's been really exciting. The people here are great. I've I've met some amazing folks. I've I've I'm getting up to speed faster than I thought I would, which is nice. and a lot of that has to do with the amount of people that are willing to help, but My my focus here at Procore is on the public sector and the public sector owners, which which is a different kind of animal from the private sector, right? They they operate differently, they have different procurement models, they have much stricter budgets, like those sorts of things. So we've gotta we've gotta get creative with that. But again, on the flip side, when you're talking about Where in the industry is there the most opportunity to still make a significant impact? That's a part of the industry that I think could really be a place where we could provide better insights, better workflows, better systems of record than what they have historically been been left with. And that's that's why I'm so excited to be here and and be a part of this particular team.
Seth: It feels like there's a future pretty soon the horizon here where an owner like I like to imagine it in a a homeowner scenario. Like let's say I wanted to build my dream house somewhere in the hills in Colorado and I could just bring my phone and talk it through, I found this site, let's take a video of it. I want a house like this with my master bedroom over here and a window that lets the sun rise and set over you know, like think through the project and then it would be like, all right, I got your your bridging documents ready for you to go out and bid it to contractors and it will already send it to contractors and receive bids and and pick the the best one, the best value and and get them signed up and then like you put a camera on site and it you can know you can just ask it, is are the beams in yet? Any issues with waterproofing, you know, and it will tell you what you need to know. so I think AI is gonna handle so many things so fast that I I I want you to kind of talk me through how you if you think you could like imagine the next five years and then like how are we gonna transition into a place where there's a lot less people needing to be involved in this? Or do you think that's just fantasy?
AJ Waters: I don't here's the thing. I don't think there's less people that need to be involved. I think
Seth: Sure.
AJ Waters: their involvement changes. Right? Like I don't think that AI is gonna make it so we go out and we fire half our staffs. That I think that's that's a a a silly way to look at it. I think we'll be able to do more with the same amount of people, right? Like that's That's what I think the the goal with tech should be to make us more efficient to where we can do more with the same. We can scale the business faster. there's there's a a part of what you explained there that I like. That's setting up the camera and just being able to ask AI what's the progress of the of the build. I think we're getting close to stuff like that with some of the the the progress input methods that we talked about earlier. There's a part of it that I'm not so keen on yet. And and let me tell you why. And that's the walk up with the camera and say, here's here's the the building I want, and I want the office space over here and this over there and that, and have it create the bridging documents. There's still a missing piece in the middle there, right? Because like I can I can verbally articulate what I think I want, but until I start to see it. I won't really know, right? Because I can I can think that I'm saying something. Like, how many times
Seth: Totally.
AJ Waters: is what you said been lost in translation by what someone else heard? And e and a AI is no different. I was actually having this conversation this morning with a with a friend of mine on on our podcast. Like he he builds custom things like this. He built this built in behind me. And he was building one for a friend. They they gave him every detail that they wanted verbally. He kind of sketched it up. They they said, Yeah, that looks great. He gave him progress photos. By the time it was done and they installed it, it ended up being not what they wanted, right? Because they didn't really see it. And so that's where I think the bridge becomes we finally get to a place where the model is useful, right?
Seth: Yeah.
AJ Waters: And and that becomes kind of that bridge of what I articulated to you is what I'm now seeing. And so augmented reality that can put that in place where I'm standing on the site and I can see the building there, but it's not, right? That kind of concept will be will be a big win before we get to anything where we're just trying to explain the project out.
Seth: You're very right. And it does make me think the more I work with AI, the more I realize how important like a feedback loop or a a lessons learned
AJ Waters: Mm, mm-hmm.
Seth: type situ like you have to enforce some kind of like a okay, that's not how I like it. Show me a new draft a little bit more like this. And you have to iterate a bunch of times and even like I I sometimes start with a prompt and iterate on it to the point where I I realize 10 prompts in. I should have just prompted with this at the beginning, but it's like the system was helping me figure out a more powerful prompt as I went on. But that feedback loop I think is extremely important. Obviously, we need human judgment involved too. I do think I'm skipping a lot when I say AI will just create the design for you. Well, I still think there's a place for engineers who know the the local soil. And people
AJ Waters: Right?
Seth: who understand the regulations in those particular areas. And then even just like letting it flat out select contractors or designers or anybody else, like there's a human judgment layer in there that we're always gonna want. I talk about it a lot. You said the one neck to choke thing. I totally use that a lot, even though it is kind of negative. It's still, it's just like people, I don't think owners of big building projects are gonna be like, yeah, I'll just let AI do it. I mean,
AJ Waters: Right.
Seth: I think we're gonna have people in front of us for a long time. So I agree with you. You talk about that too, like in these change management moments that like a lot of it has to do with finding people who kind of have a superpower for it and and giving them empowerment for it. Talk about that. Like how do you find the the advocates and what are they good at? What are people in company in companies that succeed with a rollout? They have somebody who kind of like takes it under their wing.
AJ Waters: Yeah, so there's there's two kinds of champions that I think every major successful rollout needs. Number one, they need the influential, low level, uping up and comer, right? They're usually they're probably on your your succession plan somewhere or you you see them as as an up and coming leader, but they have influence among their peers, right? So when When they say stuff, people listen. not in a way that they like beat them down, but they're likable, right? They have charisma. That that that sort of you need that sort of individual. And then you need a similar version of that at the executive level. And a lot of people think that the executive sponsor is there to like beat the hammer and drive down through the organization. This change is coming and we're gonna change, and you're all on board or else, right? But it's actually the exact opposite. The executive sponsors' number one job is to get their peers on board. Because there's nothing worse than a district or area manager that says, we're not doing that. I don't care what they say at corporate. We're going to keep doing it our way. Don't worry about it. Because that creates more trouble than anything else. And so that That executive sponsor has to get their peer group on board with the change as well, explain what's in it for their peers. And and I think that often gets overlooked, like that little piece of of that executive sponsor's influence gets overlooked. Yeah, because at that level, how many times is it competitive, right? Like my district's up against yours for bonuses and and prizes and all that kind of stuff. and and I'm I'm gonna keep doing it my way because I know that I'm not gonna slow down. You're gonna have to slow down as you implement and do all this stuff. And and so that's that's vitally important that you have champions at each level.
Seth: Hmm. That's good. Well, I'm sure I'll I'll probably bug you again in six months or a year or something and try to get you to come on and do an update. So I hope you stay tuned for that. And do you wanna plug anything? Co reach out to AJ on LinkedIn. Anything else?
AJ Waters: That's right. Yeah, look me up on LinkedIn. It's AJ Waters, Super Simple. and and I'm always willing to meet new folks in in the industry.
Seth: Very cool. Thanks, Adrian.
AJ Waters: Yeah, you bet.
