Building the Future on a Foundation of Data with Teichert Construction

00;00;00;00 - 00;00;34;11

Jason Pringle

Hi, everyone, and welcome to a special edition of our podcast. I'm Jason Pringle, a technology lead here at launch, and I have the privilege of working on some of our industry's biggest transformation stories. Today we're talking to the experts at Teichert, a construction giant that has been building California for 135 years. Clearly, they know the formula for success, and that's always included investing in their people and technology and more and more and data.

00;00;34;13 - 00;00;58;04

Jason Pringle

We're here to discuss Teichert's newest innovation, a first of its kind virtual assistant called inquiry that helps them build a smarter, build better and move faster. I can't wait to hear how they intend to use their investment in data and machine learning to propel the next 135 years of heavy civil construction. Let's get started. I'm joined by my colleague, Lynn Lighty, a senior product designer here at launch.

00;00;58;06 - 00;01;22;18

Jason Pringle

And from Teichert, we had the honor of speaking with President Mary Teichert, president of Tiger Electrical Holdings, Tom Griffith, and construction operations business manager Brian Jensen. Then take it away. Okay. First questions for you, Mary. So tell us about points in Teichert history where you've really embraced the transition or the change. So the transitions at Teichert have been starting from day one.

00;01;22;22 - 00;01;40;25

Mary Teichert

I'd say adults, Teichert history was that he was a foreman for the guy who invented expansion joints. And so he was working in that capacity on a crew in San Francisco. And then when they started building the state capitol in Sacramento, he moved up here and hung out his own shingle. So I'd say change started at the beginning, was saying, I know how to use this cool tool. I'm going to make an interesting business out of it, and I'm going to respond to what's going on in the state. So we've been making changes since day one, and I'd say it's probably never stopped.

00;01;40;28 - 00;02;01;28

Jason Pringle

That's great. Tom, I think we'll shift to you. So what what pushed this interest in bringing on the AI and the new data platforms?

00;02;02;00 - 00;02;31;14

Tom Griffith

Well, frankly, we got tired of having project failure and losing money. But, you know, we took a deeper dive at some of our project failures and learnings and the contributors, as you know, over the last 25 plus years for me have really been a lot of the same factors. Additionally, you know, in discussion with our industry peers really noticed they're having a lot of the same, you know, failures kind of from the same factors.

00;02;31;14 - 00;03;03;09

Tom Griffith

And it makes sense because this is really a human dependent business. And so as humans, you know, we're not going to be our best all the time, every day. And even the smallest mistake can take a well bid project and, you know, really increase the risk or have have some some margin erosion. And so so we reflected back and and noticed that technology was really coming to a point in other areas of the world that we felt we could leverage that in this industry.

00;03;03;09 - 00;03;43;21

Tom Griffith

And try to minimize, you know, the human failure aspect around projects. And and so that that really was one initial drive, I'd say. The other thing that really came to the forefront is our current process really was around best practice processes that we felt drove good results. It was pretty manual. We would every three, four or five years, we would refresh those practices, put them in binders, perform some training, and we realized in many of the failure areas that, you know, we have attrition.

00;03;43;21 - 00;04;04;05

Tom Griffith

You know, we have people that retire, that have a lot of knowledge, and that knowledge can walk out the door. We've realized that we hire new people outside of those training cycles and they don't necessarily get this key training. And so we really just said, you know, there's got to be a better way. So I think it's a great segway.

00;04;04;12 - 00;04;24;25

Jason Pringle

Brian, you know, how would you describe what inquiry is? Because I think that's you know what, Tom heated up. You know, how would you describe inquiry and what it's supposed to do for you guys?

00;04;24;25 - 00;04;41;23

Brian Jensen

Yeah, So I think I quit IQ as a foundation for how we kind of digitally transform our organization. You know, software that enable us pull all of the different silos of data that we have together into one spot. So the whole project team can be using the same system. It allows us to start with the bid list and go all the way to the retrospective at the end of the job to allow us to kind of capture that information, that data, that insight, and then put it all in front of the users, our our people, when they need it, to allow them to make better decisions.

00;04;41;25 - 00;05;04;04

Brian Jensen

So kind of boxes that digital assistant to move or to make us more efficient and improve our consistency of our process. Well, and Tom, how would you think that this tool is different from other data analytics tools that you guys might have considered? So we have good data organizing tools that that do collect data and categorize it for us.

00;05;04;06 - 00;05;38;20

Tom Griffith

We can run reports, perform analysis and see course corrections. What's different about I.Q. is that it takes the use of this data to the next step. It's not a passive or static system that you need to go run reports to analyze the data because IQ analyzes data for you. Now, that's based on the business ruleset. And it can catch not only mathematical errors but also makes suggestions, predict potential losses, identify when staff may be struggling in the process, things of that nature.

00;05;38;22 - 00;06;02;28

Tom Griffith

I kind of believe that it's equivalent to say a wildfire out there in the middle of nowhere, catching it in the early stages at the first sign of smoke, you know, before it's really gotten out of hand. IQ can also seek out issues for managers, giving them more time to stay ahead of the problem and work on course correction as opposed to spinning, you know, a fair amount of time trying to find the issues.

00;06;03;01 - 00;06;21;23

Tom Griffith

So this will have a significant impact in leveraging their time and capacity and help them make key decisions at the right time as opposed to more of a reaction. In the last point, I think is really key. I mean, part part of what I think makes us unique is it's not just about presenting the data, like you said, right?

00;06;21;26 - 00;06;39;17

Tom Griffith

It's it's also about triggering those insightful conversations between people who are like, given this, what do we want to talk about? Right. How do we want to talk about it? And I think even moreover, being able to take the outcomes of those conversations and weave them into future iterations of the tool so they can do even smarter things for us as we go.

00;06;39;17 - 00;07;10;11

Mary Teichert

And I think that's another really compelling aspect of the tool, which I find very exciting. Maybe back to back to you. And if you want to, anyone? Yeah, absolutely. So we can make a piece of software like inquiry without really digging in on the current state processes and diving in on the data and where we want to go for the future state and what each of you brain and in time to tell me something that you learned about teichert through this process of creating inquiry, something you maybe didn't realize.

00;07;10;13 - 00;07;29;11

Mary Teichert

Well, I'd say that one of the things I came to fully appreciate is how creative and interested in improvement our people are. One of our core values is continuous improvement, but it's one that sometimes it's easy to lose sight of. When we do projects that are more similar than different. You're over here and they we the evolution can be gradual.

00;07;29;14 - 00;07;46;14

Mary Teichert

And so seeing something that's as big of a change as this can be and having people really be able to appreciate and embrace it has been terrific. And it's a thing where I have a high opinion of our people, but you don't always see them responding to change or needing to. So watching people see and embrace the possibilities of this tool has been terrific.

00;07;46;21 - 00;08;17;10

Brian Jensen

Yeah. For me, you know, one thing that I've learned about Teichert, you know, we've been in business for 135 years plus. All right. Plus a month. Plus a month. Yeah. So we've been in business 135 years plus a month. So we we have a ton of experience under the name Teichert. I think that sometimes we feel that we still have that experience in today's state.

00;08;17;13 - 00;08;49;01

Brian Jensen

But the reality is, is that that experience and that information is in, you know, a select few people's possession in their mind. We have rooms and storage areas full of binders and information that is really not organized and not accessible. And and so really, I think that a construction companies experience is really only as good as what's available to the people that need to use it today.

00;08;49;01 - 00;09;10;21

Brian Jensen

And so learn that we haven't really had a good way of capturing all that knowledge and experience over the last hundred and 35 years. And so believe IQ will help us do that for the next hundred 35 years. I would kind of mimic Mary's point there. Everybody that we've engaged with from the taker side has been really looking for a better solution and want to help get us there.

00;09;10;23 - 00;09;33;08

Jason Pringle

And where I thought there might be resistance to change, there really hasn't been. And everybody kind of wants a better tool, needs a better tool. And we're we're pursuing that path to get it to them. So I think it's been awesome. Okay, wonderful. Bryan, we heard from Yanez a little bit about not making an oversight on the smoothness in the Caltrans bid based on inquiries.

00;09;33;08 - 00;09;54;15

Jason Pringle

Question prompts. So do you have any other examples of how employees have got some increased confidence or or missed or being prevented from making a stumble using IQ?  

00;09;54;16 - 00;10;16;00

Brian Jensen

Yeah. So I think a couple of points there. So, yeah, it gives the managers the ability to go in and look at how the estimated hours were, you know, using the tool or not using the tool. Make sure that we're having a good bid go out in the first place. So how they've answered some of the questions in IQ allows them to kind of dive in to make sure that we're having an accurate bid, go out. The historical cost and production tools and the spot check allow us to check that better and where we've been able to take off some dollars where it might have been a little higher than our historical cost, which has allowed us to win a fairly significant project.

00;10;16;02 - 00;10;38;07

Brian Jensen

So yeah, there's been definitely issues of that. And then some of those other question prompts have like checking about a sewer bypass valve prompted that estimated to go back and look at the plans again and maybe cut something that would have gone out without that that bypass valve. So, yeah, good. Glad it's working. Jason, back to you. Sure.

00;10;38;07 - 00;10;58;14

Jason Pringle

So question over to Mary. How would you see inquiry changing the way your business, which has been around for for as we've heard, quite a long time? How do you see this helping you move into the future? So being here 135 years and now five generations, we are always clear that we stand on the shoulders of people that came before us.

00;10;58;15 - 00;11;23;06

Mary Teichert

And so there are a lot of things we're doing now that were learned by previous generations of people about how to lay sidewalks, build roads, do underground infrastructure. And now the next level that we're going to be able to provide for future generations is also how to move smoothly through all of the baseline foundational aspects of asking questions and directing project work that is captured in inquiry so that people can have a higher baseline that they're starting from.

00;11;23;08 - 00;11;45;15

Mary Teichert

So instead of someone having to reinvent the wheel or rediscover what what questions one should ask in the course of different bids, all of that will already be in the system and then they can start being creative or listening to customer needs or responding to the job conditions from there. So instead of having to start where we all started, they can start from a higher plane just from having begun with inquiry.

00;11;45;17 - 00;12;17;07

Jason Pringle

Wonderful. And back over to Tom and Brian looking towards the future, what advanced capabilities, maybe including AI and ML and and other data analytics techniques do you see in the future? Inquiry I think, you know, now that we've built the foundational infrastructure, you know, to develop some AI from, you know, I think that the future functionality really is endless.

00;12;17;09 - 00;12;49;24

Tom Griffith

Few examples that come to mind. You know, we anticipate IQ being able to be assistance in, you know, our bid selection process and being able to prioritize or sort through our bid list and make suggestions on, you know, which which teams should be bidding, which projects, you know, based on where we've been the most successful or where the most risk may be or, you know, other factors, you know, and that really can apply across the board.

00;12;49;24 - 00;13;14;19

Tom Griffith

You know, which teams are best at managing, what kind of work, getting ahead of safety, you know, where are we having the most areas, challenges, you know, with safety and getting ahead of it and automating tailgate topics or trading suggestions to projects that have those items prior to doing work. So really, really in my mind, it's endless. What do you think, Brian?

00;13;14;22 - 00;13;31;21

Brian Jensen

Yeah, I would agree. I mean, I think there's now that we're collecting this data in a more systematic way, we can really start using it, and that potential is kind of endless. So from the bid schedule and what should we should go after to when we are building a project and maybe we don't identify risks that other projects have had like that can not be notified.

00;13;31;21 - 00;13;56;13

Brian Jensen

So I think there's a lot of opportunity here by capturing this data a lot more methodically that will allow us to use those future technologies again more in the future. Jason, what do you think? Yeah, I mean, I agree. I think that's one of the challenge is we had as we started operate is, you know, data lives in very different systems like bidding.

00;13;56;13 - 00;14;18;06

Jason Pringle  

Data is kind of closed up in the bidding system, right? Project execution data was over in the project management system, but we want to be able to you want it to be able to see, oh, how is this big compared to jobs that we've done? Right? And that means we have to plan. We have to find ways to collect those pieces of data across the systems.

00;14;18;06 - 00;14;40;05

Brian Jensen

But now that we've found consistent ways to do that algorithmically, not just by convention, I think it's opening a lot of doors for us and also providing those we want to pull in other systems. We've kind of for scale that same the same challenges, but we've done it once and we have a bit of a roadmap for how to do it as we go forward.

00;14;40;07 - 00;15;03;14

Brian Jensen

Yeah, I think in an example I can think of is we did spot check, right? Spot Check is a wonderful example of this, right where we want to surface using historical analysis of jobs. Right. What are some problems I might have in a bid that's sitting in front of me? And you know, I think as we first jumped into it, we ran into all sorts of problems like, well, how do I find jobs that are similar to this bid?

00;15;03;19 - 00;15;21;18

Tom Griffith

Right. What what does it mean? We had to talk about what does it mean for a job to be similar or a bid to be similar? Is it just where the work is to the type of work? You know what? What are those axes of similarities that we wanted to then go find differences where things are outlined and this one's not like those because it seems to be at odds with each other.

00;15;21;21 - 00;15;57;13

Mary Teichert

You know, I want to find things that are similar to look for the differences. And that took a little while for us to all wrap our heads around it, I think. But now that we've kind of got into that groove, I think it opens up a ton of doors for where you can go. Yeah, I agree. I mean, the safety net aspect of some of the features as well as the the guidance piece really does well, it's kind of like having a full time trainer with everybody, though younger or newer staff can can get that guidance, can get those question prompts, can get the spot check.

00;15;57;15 - 00;16;32;23

Jason Pringle

So in this in this business that is again very human dependent and a small error which which can happen can turn into the wildfire, so to speak. So the safety net is comforting for me. So, Mary, you had mentioned that creating a better experience. You know, you mentioned creating a better experience. And so, Lynn, I have a question for you, which is, you know, as our human experience expert, can you give us some examples of how Teichert is focusing on that human experience with this project?

00;16;32;25 - 00;17;02;22

Lynn Lighty

Yes, absolutely. I'd be happy to. So access to the the subject matter experts and users of inquiry has been very important in that entire It's been very gracious about giving me and the other designers the time and access to them to understand really how they're doing, the work that they're doing, how they're moving between the tools, they're existing, how we are going to put inquiry into that process in a way that is useful and not disruptive.

00;17;02;25 - 00;17;24;13

Lynn Lighty

Being attentive to the ways they are thinking about the data, how they organize it and label it, ways to smooth out the task. We've been working quite a lot with the estimating team to tune the data. I know Jason who's been involved in that, so that the results, our results out of the data analysis that they have confidence in.

00;17;24;13 - 00;17;51;18

Lynn Lighty

So we have review sessions with the estimated hours to update and tune that algorithm we have been working hard to emulate the kind of data capture that you we have had in Excel sheets in the past and bring that data into the system so it can be live, but make sure when we are doing a screen flow or a layout, we are attentive to as little friction in that process as we can.

00;17;51;18 - 00;18;16;08

Lynn Lighty

So we want the data entry to be easy and seamless. We've had quite a lot of work around autosave features and moving between screens, saving the data, being able to provide photo uploads and other aspects to help give the estimate or the project manager enough latitude to provide the level of information they need to. So a big complicated job.

00;18;16;08 - 00;18;40;01

Lynn Lighty

They can expand the template and hold as much data as they need a smaller subdivision job that's later. They can do a leaner version so we don't lock them into a process that doesn't quite fit. One size fits all has been one of the kind of rubs in the older processes. So we've been working to make the system flexible enough depending on the job size.

00;18;40;03 - 00;19;15;06

Lynn Lighty

And I've got a question back for you, Jason, too. You touched on the disparate systems before, so that's a challenge a lot of other companies have when they're trying to embrace a digital transformation. It's not at all uncommon. So I guess maybe back to Tom and Brian or even Jason, if you've had other projects, are there parts of the business that were more ready for this transition than others when it comes to this, like the disparate data or kind of coming into the HQ?

00;19;15;08 - 00;19;34;14

Tom Griffith

I don't I don't think so. I think the whole organization was ready for it, really. I mean, I think we've made the process easier by kind of connecting those dots together versus having them over here or in Excel where we had to do it multiple times because the estimate or came up with some risks and then the PM has to come up with the same risks or different risks and then interim from this Excel sheet into our ERP system or another Excel sheet.

00;19;34;14 - 00;20;05;10

Tom Griffith

So it just really reduced the redundancy and then it made the process easier for everyone involved. So I think everybody's been really embracing of it and I think even the technology side or the, you know, often you get into trying to dig into products data and you hit a lot of walls, right? And I think every time we've talked to anybody on the team about how do we go get this data, they just jump right in and say, oh, well, either I know how to do it, here it is, or let's go figure it out.

00;20;05;10 - 00;20;33;17

Tom Griffith

Right. And and I think that's one of our keys to our success as a team is for Mary and anyone else in the room who wants to throw an answer out, which is, what would you say to other construction giants who are deciding to invest in a similar or deciding how to invest in a technology transformation? I would say to other construction giants that it's a challenging thing for a construction company to embrace technology.

00;20;33;23 - 00;21;06;15

Mary Teichert

There are some study that says we are the second least technology oriented industry in this country. We're right about hunting, so we're just not an organization in a type of industry where that's the thing that we lead with. Most of the time we're building things that we like to see and we're working on projects. So taking this step into 21st century capabilities and saying that in addition to ground engaging tools, one of the essential things we need to do our jobs has to do with modern technology and with this kind of tool is going to be making us able to take our business to the next level.

00;21;06;18 - 00;21;25;08

Mary Teichert

And partly it's because it helps us leverage the incredibly smart, capable people who are getting into this line of work. So if you look around at the the excellent resources we have both outside in the field and then also in the office doing all of the management and estimating and work we do, part of what they're looking for is to have tools that make their job simpler.

00;21;25;10 - 00;21;55;12

Mary Teichert

Everybody at this point, since the time they were toddlers, has been familiar with technology. So the generations around us are shifting rapidly and making a construction company that people feel like they're excited to work for and where they have the tools they need to do their jobs. An essential piece of that is going to be increasingly about technology and what we have built here is pretty cutting edge because it captures both what we had previously understood and our wisdom about construction work, as well as the new that all of the capability is that new technology can bring to the table.

00;21;55;14 - 00;22;19;09

Mary Teichert

So I would say to other construction companies that are looking for ways to both engage and develop the workforce they have and to move their work to the next level, that this tool is something that provides that kind of platform. Yeah, for me, I'd add to that advice or what I would say to other other construction companies looking to make a digital transformation is one you got to start somewhere.

00;22;19;09 - 00;22;45;26

Brian Jensen

And to build this foundation of data, a different kind of data in a different way to look at the data you have, have the vision, have the discipline to stay the course because you've got to start somewhere and it will eventually pay off. And so for us, you. Who doesn't want an instant solution? But but realizing that it doesn't come overnight.

00;22;45;26 - 00;23;14;01

Brian Jensen

So, you know, this has been kind of an iterative build and it just keeps getting better. It just keeps getting better at that. Those are fantastic answers. And I could not ask for anything better. I love that you talked about investment, that you talked about the employee retention. That's a big topic right now to me that I would have loved to hear earlier when we talked about comfort insight in the people who are using it is about that workforce retention.

00;23;14;02 - 00;23;48;18

Brian Jensen

Like, how could one of you respond a little bit about how reducing stress on the workforce, how it might be something that. So you're talking about ways it improves the job for the people that work here. Exactly. I'd love to hear more about that. So one thought on that, on ways that inquiry is going to improve the work for the people that work here is that it allows everyone to use their creativity and their construction capabilities to address the more interesting elements of how do we meet this customers need and how do we respond to the unique elements of a project without having to reinvent the wheel.

00;23;48;24 - 00;24;08;00

Tom Griffith

Here's the basic questions that we've asked on the last 300 projects as well. And here are the things that someone who's been doing this longer than I have may also know. All those questions are captured in inquiry already so that folks are able to really use their creative creativity and intelligence and capabilities on stuff that's unique to the project that they're on.

00;24;08;02 - 00;24;23;12

Tom Griffith

So I'd say in terms of general job satisfaction, people are able to really do the work they love to do, which is meeting customer needs and building great work, and they don't have to be spending a lot of time trying to re discover what are the things that they were supposed to have analyzed and looked at coming into the place that they are now.

00;24;23;14 - 00;25;00;26

Tom Griffith

So I think it really helps them elevate the work that they're doing to another level, which is satisfying to good people. Good people want to do really interesting work and this lets them do that. Yeah, I also think that to Mary's point around the system, being able to do a lot of kind of the science, so to speak, and in allowing our employees to focus on the the art part of the process, you know, we are hoping that that it will create more time and capacity to get creative, as Mary suggests, and also kind of find that work life balance, that that is part of our core value.

00;25;00;26 - 00;25;23;15

Tom Griffith

So it makes people want to come to work every day because they know they can come home and see the family that they are feeding with all this good work. It seems like another thing we've you mentioned art and science, and I think it feels as though and I'm so I'm curious your guys's reaction like that as things that we thought might have been science has actually turned out to be a bit more art and sometimes vice versa.

00;25;23;15 - 00;25;41;24

Tom Griffith

As we dug in to a particular way, somebody looked at something. And I'm just curious like what your guys's reactions to that might be. So one story about things that we used to think were art that turned out to be science. I've told a couple of times about back when we first started having an organized work breakdown structure.

00;25;41;27 - 00;26;00;25

Tom Griffith

Prior to that point, estimate in different offices would number their estimates in a way that made sense to them. And so when we first got together and said we ought to really have a work breakdown structure in common, there was some pushback and people felt like, But then we I want to be able to be creative. And so it was a bit of a change of mindset to say, Well, we want you to be creative too.

00;26;00;25 - 00;26;27;05

Tom Griffith

But how about instead of how your estimate is number, do you focus all that energy on like where's a good water source for this job? So trying to shift people's thinking about what is involved in the art and what's involved in the science as part of the opportunity that the software creates, because then we can actually say some things that might have felt like art are actually pretty intricate, subtle science, and you can ask the same questions that applied and you can do some pattern recognition on jobs that we've previously had.

00;26;27;07 - 00;26;49;20

Brian Jensen

One benefit of being the kind of contractor we have we are is at any given moment, the company has between 203 hundred active jobs and so there's a lot of wisdom being captured on all those jobs and saying that a software as intelligent with algorithms and analytic capabilities that IQ has makes it possible to look at for patterns across most jobs that an individual person might not even be able to see.

00;26;49;22 - 00;27;12;18

Jason Pringle

And so combining that level of science with what really is art in terms of what does the customer really want, what's the most important thing to them, kind of takes our business to a new level. Also to you mentioned you have a few hundred jobs running at any one point in time. Could you give a just an estimate on what a typical dollar size is for one of those jobs?

00;27;12;21 - 00;27;32;01

Brian Jensen

Right. Well, it's a good question. I'd say we have like a bell curve of job sizes. Probably if you took an actual average, it would be around $5 million. But we have them as small as 500,000 and as large as 150 million. So it really depends on the job. Our sweet spot is probably in the like 15 to $25 million range.

00;27;32;03 - 00;27;46;10

Brian Jensen

Most of our jobs, our goal is to not beat the company on any one specific job most of the time. So we spread it out quite a bit and a lot of our customers have outstanding work to do that's in that size range. But if we get too small, then it really is a better fit with maybe a smaller contractor than we are.

00;27;46;10 - 00;28;06;28

Mary Teichert

So trying to find that ranges where we spend some time. And that's one of the reasons Beat selection is so important, because trying to figure out it's not as simple as saying this job fits us because it's the size in this type of scope. There's a lot of considerations, such as, you know, ranging from the customer to our resources to what else is going on with our strategy that makes that selection.

00;28;07;01 - 00;28;26;23

Mary Teichert

One of those things that has both art and science involved in it, which is a place that I'm really glad we've got tools like IQ to help us really figure these things out, what matters and what's going to make a difference to us. Yeah, I would just say thank you. It's been a pleasure working with you guys so far and I'm absolutely looking forward to a more productive time together.

00;28;26;25 - 00;28;33;10

Jason Pringle

We've enjoyed spending time talking very much.

‍