Back to podcasts

The pro PM: Avoiding the little mistakes that cost you big

In this episode of The Cost Code Show, Ryan Gilmore sits down with Matt Graves — seasoned construction project manager and creator of Construction Yeti — for a practical conversation on what separates chaotic projects from well-run ones. Drawing on experience from both the subcontractor side and the owner’s rep side, Matt breaks down the project management fundamentals that trade contractors can’t afford to overlook, from building repeatable processes and improving field-to-office communication to capturing change orders before they become lost revenue.

The conversation also digs into one of the biggest challenges facing growing subcontractors: knowing when to rely on process, when to bring in technology, and how to build a business that can scale beyond one owner wearing every hat. Matt shares honest, field-tested advice on job costing, labor tracking, document control, and why the best systems are the ones that actually make life easier for the people doing the work.

Whether you’re a project manager trying to tighten up your workflow, an owner looking to grow your operation, or an accountant supporting construction clients who need better visibility into their jobs, this episode is packed with grounded insights you can put to work right away.

Resources & Takeaways

  • Learn why standardized project management processes are critical for subcontractors that want to grow without creating chaos.
  • Understand how stronger field-to-office communication can prevent delays, reduce friction, and improve project execution.
  • Discover how to capture and present change orders in a way that protects margin and speeds up approvals.
  • See how better job costing and labor tracking can help contractors spot problems early and course-correct before profits slip away.
  • Identify where technology can genuinely support a construction business — and where process still matters more than software.

Listen

Watch

Read Transcript

Ryan Gilmore:

Hello everyone. Welcome to The Cost Code Show presented by Knowify. This is a podcast where we focus on helping trade contractors run and grow their businesses.

Today I’m excited to welcome on Matt Graves. Matt is a seasoned construction project manager, and he also runs Construction Yeti, a website and newsletter dedicated to helping the next generation of construction professionals.

Matt has a ton of in-the-field experience that he’s going to share with us in today’s episode. How are we doing today, Matt?

Matt Graves:

Doing great. Thanks for having me.

Ryan Gilmore:

Of course. Matt, for folks who don’t know you yet, can you give us a quick look at your background and the kind of work you do in the construction world?

Matt Graves:

Yeah, I’ll do a long story really, really short.

I went to Texas A&M, got a civil engineering degree, and decided I did not want to be an engineer. So I got out of school and found a job at a subcontractor. I had a couple of stints at different subcontractors and really learned that game. Then I jumped from there to the owner’s rep side.

I haven’t been with a general contractor, but I’ve kind of seen them from both directions, practically speaking. Now I think that’s given me a really holistic view of the industry. I always thought, man, if I went back to the subcontractor side, I’d have a much better grasp of things because now I understand the “why’s” behind a lot of what happens.

I’ve been on the owner’s rep side now for about seven years and realized there’s a lot that people think they know—or have misunderstandings about. There were things I thought I knew and had misunderstandings about too. Over time I’ve gotten a broader worldview, which led me to launching Construction Yeti—trying to share insights, resources, and tips and tricks that I’ve learned over the last decade and a half of working in the industry.

Ryan Gilmore:

Awesome. That’s a good lead-in to my first question. If you think back to your first few years in the industry, what’s something you wish you had learned earlier about managing a job or running a crew?

Matt Graves:

There was so much stuff that I didn’t know that I didn’t know. I was 24, 25 years old, fresh out of school, and I thought I knew what I was doing. In hindsight now, I had no clue what I was doing.

But I was running crews. We were a federal subcontractor. We had crews from the East Coast to California. My phone would start ringing early in the morning as the East Coast guys got to work and stop ringing late in the evening when the West Coast guys got off work.

Things I wish I knew back then are probably things that only come with experience—the falling down and scraping your knees kind of lessons. But one big thing is just having patience. Everybody gets tense, but at the end of the day, I’ve never been on a project that didn’t get finished. There’s a lot of screaming and yelling and drama along the way, but you’ve got to take a breath and just tackle what’s in front of you—next thing, next thing, next thing. It’ll get done.

Ryan Gilmore:

Nice. So Matt has given us a good segue to the topic of the episode. He’s going to walk us through the project management basics that every contractor should nail.

Jumping off from that: what are some of the small mistakes you see contractors make that might seem small at first, but actually turn into big problems later on?

Matt Graves:

Really, I’m talking about subcontractors here, because I haven’t been with a GC. On the subcontractor side, a big one is not having project management processes specifically for the subcontractor—getting really niched down there.

I’ve seen this in a couple of different companies, and even now on the owner’s side. You work with a subcontractor on one project, then with the same subcontractor on another project, and it’s a completely different experience. They do things differently, and you’re looking at it like, “You guys aren’t uniform at all.”

So if one project manager gets hit by a bus or whatever, the next person steps in and there’s nowhere to pick up from. As a company, you want to be consistent so you can improve and train your people.

I think not having a project management manual that you actually train your folks on is a big miss. It sounds like “just another rulebook,” but it should be a living document that you actively train people on.

Things like: here’s how we do scheduling. Here’s how our project managers interact with our field crews. Here’s how we do look-aheads. Here’s how we support them. Even change order management—because I’ve seen subcontractors who are really good at capturing every little change, and others who let million-dollar change orders almost go by without capturing them.

There’s a fine balance—you don’t want to nickel and dime your client, but you do want to make sure you’re made whole.

Ryan Gilmore:

Cool. I like the idea of this project management “rulebook.” If you had to list out some of those core PM rules for a subcontractor that doesn’t have a proper PM department, what would be some of the big, high-impact ones?

Matt Graves:

The best ones are the ones that hold your hand through the whole lifecycle—through the contract process and throughout construction, all the way to closeout.

Maybe it doesn’t have to cover estimating if your project managers get a handoff, but it should clearly define what turnover day looks like. You won the project, congratulations—you’re the project manager. What does turnover look like from estimating? What does that estimate look like? How does that roll into your budget and your schedule? How does that roll into helping the field foreman or superintendent build out their work plans?

Maybe you help build those plans, maybe you don’t—but you should at least define it. Then walk through: how do you do buyout of materials and subcontractors? How do you manage submittals, RFIs, changes, and so on—all the way to closeout documentation.

Ryan Gilmore:

Nice. Based on that, I want to give you a kind of lightning round of areas that are common pain points for our customers—mostly trade contractors and subcontractors.

Field-to-office communication is a big one, which you hinted at earlier. What are some practical ways a trade contractor or subcontractor can improve communication between the field and office? This doesn’t have to be tech—it can be process, and then maybe some tools that can help.

Matt Graves:

Ultimately, we’re all wowed by the tech—me included. We think, “How can I digitize this? How can I automate that?” But at the end of the day, you’ve got to build a relationship with the field.

You’ve got to get that face-to-face time. Sit down with them. When they have a problem in the field, walk out there and look at it with them. Do weekly look-ahead meetings. Sit down and ask: where are you going next week? What do you need? What do you need from me, as a project manager, to clear obstacles out of your way?

Maybe there’s an RFI holding them up in an area. Maybe a change order is impacting their work and they need you to push the GC to get it approved so they can move forward.

So really, it’s that face-to-face relationship-building. For example, when I was with one subcontractor, we were on a project with a 40-foot combo conex—12 feet of office and the rest storage. There were four of us crammed in there with tiny desks: two superintendents, a project manager (me), and an assistant project manager.

We were tight—if I pushed my chair back I’d hit the guy behind me. But we’d all go in there and eat lunch together, watch dumb YouTube videos, laugh, and hang out.

We were in a job costing meeting once, and the president of our company said, “You guys have the tightest project team between the project management and field crew of any of our jobs. What’s your secret?” I said, “I don’t know, man—we’re just kind of buddies. We laugh at stupid stuff all day and we’re basically sitting in each other’s laps.”

On other projects, the project manager thought he was “the” project manager. He had two separate trailers: one for the office and one for the field. They were forced to have their weekly meetings, but that was pretty much the only time they talked.

So even with all the AI and communication tools we have, it really comes back to basics: being in the mud with them, seeing eye to eye, and building that personal relationship.

Ryan Gilmore:

That’s great advice. No need to be adversaries—you’re all on the same team. At Knowify we have this idea of “one team,” where we try to bring different groups together to work across teams. So that really resonates.

Another big pain point you mentioned is change orders. These seem to be a headache across the industry, whether you’re a GC or a subcontractor. You’ve got a lot of experience with them. What are some common mistakes you see, and what does a better or ideal process look like, if that’s achievable?

Matt Graves:

One huge mistake is just not capturing them.

Stuff happens all the time. If you’re a subcontractor and you bid from plans and specs, the best training I got was: keep a change order log. If anything comes up that could possibly be a change, note it in your log. This is your internal log—you’re not sharing it with the client all the time—but you’re capturing it.

A tiny little thing can quickly spiral into something much larger. If you capture it early, it stays front-of-mind. If it does spiral, you have that story ready to tell when it’s time to have the conversation.

Then, when it comes to pricing the change order, I see big differences. I thought I was decent at putting together change orders on the subcontractor side—good enough to sell them to the GC, who then had to sell them to the owner.

Now, on the owner’s side, I review change orders all the time—from the GC plus their subcontractor backup. The ones I can basically just check off and approve are the ones that tell a story.

Most contractors we work with are legitimate. But if I see what looks like a minor issue and they’ve got 200 man-hours listed with no explanation—“labor: 200 hours; materials: $1,500”—I don’t know what to do with that. I start pushing back and digging deeper. It becomes painful for everybody. I don’t want to do that, they don’t want to do that.

But if you tell the story—“It’s going to take a crew of three guys two days to do step one, then the same crew three days to do steps two through five,” and you add it all up—suddenly 200 hours doesn’t feel crazy. It makes sense. You don’t have to question it.

You’ve got to remember: the GC has to take it and sell it to the owner. The GC is a little more removed, the owner’s rep is even more removed, and the owner’s rep may have to take it to the actual owner. If you can paint the picture clearly in black and white, it moves much faster.

I make a lot of memes online about “fluffy” change orders. I get comments like, “That’s my price, take it or leave it.” Those folks don’t understand their contracts. On any commercial project I’ve ever been on, you’re required to provide backup on your change orders.

If you can explain why it’s 200 hours—with tasks, crew sizes, and durations—you’re doing yourself a favor.

Ryan Gilmore:

Yeah, that’s great. I was talking to another guest about pricing and he said, “Selling is part of pricing.” And selling is basically telling a story, which fits perfectly with what you’re describing.

I want to switch over now to visibility into project costs—cost awareness. You mentioned job costing earlier. What’s something smaller contractors or subcontractors can do to get better visibility into how a job is going before it’s too late to fix the problem?

Matt Graves:

Cost coding.

I’ve seen contractors who basically don’t cost code anything. They bid it, then go do the work, and at the end of the project they wake up and ask, “Did we make any money?” They might have a high-level labor, material, subcontractor, and equipment budget, but they’re not tracking at a meaningful level.

For subcontractors, labor is your biggest risk. You can lock in subcontractors early, lock in equipment, and mostly lock in material, but labor is where your risk lives. You thought something would take 10 hours and it takes 20.

So you need to break labor down into actual activities, then track against those. If you just say, “We’ve got 2,000 man-hours on this job,” that doesn’t mean anything. The number is so big it’s useless.

Better is to take your labor estimate, break it down by activity, and then have your superintendent or foreman use that to build out work plans. For example: to do one floor, you’ve got 100 man-hours—which is a crew of four for X days. You build milestones out of that.

Then you can check: are we trending ahead or behind? If you’re way over on labor, you can ask, “What can we do to fix it?” instead of wondering at the end of the job.

Ryan Gilmore:

Nice. That’s great advice.

I want to pivot now into big-picture, mindset stuff. You’ve got a lot of experience as an owner’s rep and as a subcontractor. You’ve seen how businesses grow and evolve over time.

What are some major revenue or business milestones for a subcontractor? And how have you seen business needs change as companies hit those different milestones?

Matt Graves:

That’s a good question.

Almost every subcontractor starts with a person who got tired of “working for the man.” I always picture a plumber for some reason. He’s a great plumber, he gets tired of working for someone else, so he starts his own company.

At first, he’s the plumber plus project manager plus business owner. He has to do all the marketing and maybe recruits his wife or someone to do the accounting. He’s wearing all these hats.

As he grows, he picks up a couple of projects, promotes a strong field person to foreman, and he’s not physically on the job every day anymore. But he’s still doing all the project management and everything else.

I used to think in terms of revenue milestones—$5M, $10M, etc.—but with inflation and everything that’s shifted a bit. So now I think more about organizational milestones.

The growth steps I see are tied to how the org chart fills out. If that master plumber is still doing all the estimating, all the PM, all the things, it’s a small company. As you start developing departments, processes, and systems, you move up a level.

If one person is doing all the estimating and project management, they can only handle so many projects—maybe 5–10 depending on size. Once you start building out a real team and processes, you can scale beyond that.

Ryan Gilmore:

I think that’s great. So if someone is that “master plumber” doing everything themselves, and they’ve now got a little crew, someone doing the books, maybe a PM/estimator helping, but they’re still in the field doing most of the work—how would you coach them to get to the next level?

Not putting a revenue number on it, but moving from “small” to “more established.”

Matt Graves:

I’d start by shedding the work you hate.

If you’re the master plumber and you hate estimating, hire an estimator or promote someone into that role. If you hate dealing with an angry GC who wants updated schedules and fifteen change order prices, hire a project manager.

If you’re wearing ten hats, you probably hate five of them. Maybe you love being in the field and running crews—lean into that. Shed the rest.

Hire or promote people you trust to cover estimating, PM, maybe business development. If you’re still doing all your own BD, consider a dedicated BD person. That frees up your mind to go deeper on what you’re good at and actually enjoy.

Ryan Gilmore:

Nice. I want to talk about tech now. I know you’re big into construction technology, and it can be rocket fuel for a smaller company—helping them do more with less.

When a contractor has the basic project management blocks in place, when does it make sense to bring in tech? What are some signs they’re ready to add software to the equation?

Matt Graves:

I think earlier than people might expect.

I said earlier that tech isn’t the solution, but as you develop your processes, you really need to figure out what tools those processes will live in. I’ll use Procore as an example since everyone knows them. If you love Procore and want to use it—and you can afford it—get it early.

You’re building your company around a process. These software tools become the way things get managed. If you’re writing a PM manual, you need to know what tech you’re using. Every platform has its nuances. You can swap later, but it’s painful.

If you’re doing everything in Excel and on paper, and five years in you decide to switch to a platform, it’s a harder transition. In my day job, we’re looking at switching project management software. Our contract isn’t up for another year, but we’re already planning because we know we’ll probably run two systems in parallel for a while. All new projects go into the new system, old ones stay in the old system. Nobody wants to do that—it’s a big headache and you have to rewrite processes.

So I’d start as early as you can reasonably afford it, so you can develop your team into following a consistent process and toolset.

Ryan Gilmore:

Where in the PM process do you think tech has the biggest impact—especially for that smaller subcontractor trying to take the next step? What are the two or three areas where tech is likely to help the most?

Matt Graves:

First, document control.

That could be as simple as a well-structured Google Drive or as sophisticated as a full PM suite. But you need a consistent file structure: here’s where RFIs go, here’s where submittals go, here’s where change orders go—for every project.

Second, accounting.

Accounting almost has to be software-driven. You can’t realistically run a business on the back of an envelope. You need an accounting system, and ideally it talks to your project management system so you’re not doing a bunch of manual double-entry.

Third, AI and automation.

AI is the big buzz right now. I think understanding where it can help—especially on busywork—is important. Not every AI tool moves the needle. A lot of them are “95% right,” and that last 5% can be a pain because you have to hunt down the mistakes.

But at least keep it on your radar. Maybe in six months or a year something jumps from 95% to 100%, and if you’ve stayed plugged in, you can adopt it quickly.

Construction is still a very human, face-to-face, handshake industry. But there’s so much admin and logging and repetitive work. The more of that you can automate and bake into your processes, the better.

Ryan Gilmore:

That’s great. I’ve done a fair amount of experimenting with AI and I’ve had similar experiences. It’s incredible until it’s not—and then you have to figure out what went wrong, and sometimes it would’ve been easier to just do it yourself. But that’s how progress goes.

Matt Graves:

Exactly. I started making these Construction Yeti character videos when Google’s Veo model came out. You see the promos and think, “This is amazing. You type a prompt and it just makes a perfect video.”

Then you use it and realize, “This is hard.” You’ve got to figure out how to prompt it, it’s not consistent, and it only does eight-second clips. To make a longer video you have to chain scenes together.

The other night it just refused to generate audio for one scene—twice in a row. The third time it worked, with the same prompt. So, yeah, it’s changing the world, but it’s not an overnight magic wand.

Ryan Gilmore:

It’s not quite magic yet—but it’s getting there.

Matt Graves:

And it’s getting there fast.

Ryan Gilmore:

It’s a good point. In construction tech specifically, I’ve been in the space for about three years and it feels like the number of tools has tripled.

In your opinion, what separates the tools that actually help from the ones that don’t?

Matt Graves:

The ones that help actually do something meaningful for the user.

That sounds obvious, but you’d be surprised. I’ve done a lot of demos over the last few years just because I’m interested. I’ll talk with founders who have poured their heart into something for two years. They’re excited, they show me how they’re pulling data from all these places, and then it spits out a dashboard.

The dashboard will tell me that on average we’re late on RFIs by 7.4 days, or that our change orders are slow. I don’t need a fancy dashboard to tell me my project is behind. Everybody already knows that.

So I’ve had to break a few hearts and say, “This doesn’t actually do anything for the people doing the work.”

Don’t implement tech just because it’s shiny. It’s hard to implement, hard to get buy-in. You’ll never get buy-in—especially from the field—if it doesn’t make their job easier.

I’m a big proponent of tech and AI, but if it only exists to provide a report to executives or to squeeze more productivity out of the field, you’re not going to get adoption. It has to make life easier for project managers and field crews. They need to see a benefit.

Ryan Gilmore:

Totally. You’ve got to solve real problems or there’s not much point in existing as a product.

We’re getting toward the end of the interview, and we like to keep the podcast very actionable. If someone’s listening and they want to improve one thing about their project management process this week, where would you recommend they start?

Matt Graves:

It depends on where you sit.

If you’re a director of project management, ask yourself: are all of your people rowing in the same direction? Are you training them properly? Do a self-audit.

Look at how your teams actually use your systems. Maybe you use SharePoint or Google Drive. Is there a standard project folder structure, or does everyone just create their own random folders? Is one PM really good at capturing change orders and another terrible at it?

You’ve got to get a handle on that and at least get people moving in the same direction.

If you’re an individual project manager or coming up through the ranks, my advice is: stay curious. You’re never going to know everything. I’ve been doing this a long time, on both sides of the table, and I still ask “stupid” questions every day.

I’m constantly saying, “I don’t understand—what is that? What are you talking about?” Just keep asking questions and keep learning.

Ryan Gilmore:

Awesome. As you all can tell, Matt has a ton of experience and expertise. One of the ways he shares it is through his newsletter, Construction Yeti.

Matt, can you talk a little bit about that and let folks know where they can subscribe if they want to follow your work?

Matt Graves:

Yeah. I started Construction Yeti about five years ago. It started as a blog and then morphed into a Substack newsletter.

The idea was: there’s not enough mentorship in the industry. There aren’t enough good resources for people who want to stay curious and keep learning. The newsletter is the Construction Yeti Substack, but the weekly email is called “Construction Curiosities.”

It started as, “Whatever I’m curious about this week, whatever I’m learning, I’ll share it.” Over time it’s taken on more of a mentorship tone. I’ll do deep dives into things like delivery methods—stuff I didn’t really understand when I was on the subcontractor side.

For example, I was working on CM-at-risk projects but we had a hard-bid contract to the GC, so I didn’t think much about the GC’s relationship with the owner. Looking back, if I’d understood that relationship better, I could have been a better partner and understood their pain points more.

So I do deep dives to explain those kinds of things. But I also try to make it fun—people get enough boring white papers and dry articles. There’s a meme every week, I sprinkle in GIFs, and a lot of AI-generated Yetis.

My philosophy was: nobody wants to read another boring white paper. So I send it on the weekend, when things are slow. Saturday or Sunday morning, you’re having a cup of coffee—it’s an easy, actionable read that gets your brain going without putting you to sleep.

Ryan Gilmore:

That’s awesome. We’ll have links to where you can sign up in the show notes when we publish the episode.

Matt, I think that’s a good note to end on. I really appreciate you joining us and sharing your knowledge with the audience.

To everyone watching or listening, thanks for tuning in. If you need help managing your projects and finances for your trade contracting or subcontracting business, check out Knowify at knowify.com.

We’ll be back with another episode soon. Cheers.

Related Resources