Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How a Construction Firm Cut Project Overruns by 50 Percent
Transcript
- Lucas: If you work in any industry where you're managing a project — construction, software, event planning, even a kitchen remodel — you've lived this: the timeline slips, the budget creeps, and suddenly the thing that was supposed to take three months takes six and costs forty percent more. Luna: Right. And there's usually a lot of finger-pointing. The client changed the scope, the subcontractor didn't show up, the materials were late. Lucas: Exactly. The finger-pointing is almost always the same, regardless of the actual cause. So today I want to talk about a mid-sized construction firm — let's call them 'BuildRight' — that decided to stop blaming and start measuring. And by doing that, they cut the number of projects that ran over budget or over schedule by more than half. Luna: Half? That's a dramatic improvement. How big was this firm? Lucas: About a hundred and fifty employees, doing mostly commercial interiors — office fit-outs, retail spaces, that kind of work. They were doing maybe twenty projects a year, each ranging from two hundred thousand to two million dollars. And their track record was pretty typical: about sixty percent of projects finished late, and almost half went over budget. Luna: Ouch. So what did they change first? Lucas: They scrapped the traditional Gantt chart. Not entirely, but they stopped using it as the primary tool for coordinating the team. Instead, they adopted something called 'pull planning' — a concept from lean construction that's been around for a while but is still surprisingly rare. Luna: Pull planning — is that like the opposite of pushing tasks forward? I've heard it in manufacturing contexts. Lucas: Exactly the same idea. Instead of the project manager creating a master schedule and then pushing it down to the trades, pull planning starts with the end date — say, 'the tenant moves in on September 1' — and then works backward. Each trade says what they need from the previous trade to finish on time. It's collaborative, not top-down. Luna: So the electrician says, 'I need the walls closed by this date,' and the drywall crew has to commit to that. Lucas: Exactly. And they make those commitments in a room together, on a big physical board with sticky notes. It sounds low-tech, but that's kind of the point. When people write down a promise in front of their peers, they're far more likely to keep it. Luna: That's the social contract part. So how did BuildRight implement this? Did they just roll it out across all projects at once? Lucas: No, they started with one pilot project. The superintendent on that job was a guy named Ray — a twenty-year veteran who was famously skeptical of any new system. He had a Gantt chart tattooed on his forearm, I'm kidding, but he might as well have. Luna: I can picture him. So how did you convince Ray? Lucas: They didn't try to convince him, actually. The operations director — a woman named Priya — told Ray, 'Look, on this one project, I want you to try pull planning for the first three weeks. If you hate it, we go back to your Gantt chart. No hard feelings.' And she assigned a young project engineer to facilitate the weekly pull-planning sessions. Luna: That's smart. Low risk, high autonomy. What happened in the first three weeks? Lucas: The first week was chaos. The trades were confused, Ray kept interrupting to say 'this is a waste of time.' But the engineer kept them focused on the sticky notes and the 'percent plan complete' metric. That's the second key piece — they started measuring how many of the commitments made in the pull session were actually completed on time. Luna: So instead of measuring whether the project was on schedule overall, they measured whether each trade did what they said they would do, week by week. Lucas: Right. And that's a completely different conversation. When a trade misses a commitment, you don't yell — you ask, 'What stopped you from finishing?' And you dig into the root cause. Was it a material shortage? Unclear specifications? Another trade not finishing on time? Luna: That's the 'five whys' approach, but applied to scheduling. Lucas: Exactly. And what they found was fascinating. About seventy percent of missed commitments were caused by things that could have been prevented with better coordination. For example, the drywall crew couldn't start because the MEP rough-in wasn't done — but the MEP foreman didn't know the drywall was supposed to start that week. The schedule said it, but nobody had communicated it. Luna: So the pull planning forced that communication. The MEP foreman had to say, 'I need three more days,' and the drywall foreman could adjust. Lucas: Yes. And after three weeks, Ray's project was actually ahead of schedule — the first time that had happened in years. Ray, to his credit, became a convert. He started leading the pull sessions himself. Luna: That's the kind of story that makes change spread. So what were the actual numbers after they rolled it out across all projects? Lucas: They tracked it for eighteen months. Before pull planning, about sixty percent of projects were finishing late. After, it dropped to twenty-eight percent. Cost overruns — defined as more than five percent over budget — went from forty-five percent of projects to twenty-two percent. And here's the kicker: the average project duration actually shortened by about fifteen percent. Luna: So they were finishing faster, more on time, and more on budget. That's a triple win. Was there any downside? Lucas: The main issue was the time investment up front. Pull planning sessions take two to three hours per week, and you need all the key trades in the room. That's a real cost. But BuildRight found that the time saved downstream — fewer fire drills, less rework — more than made up for it. Luna: And I imagine the culture shift was also a challenge. Not every superintendent is going to be as open as Ray was. Lucas: Absolutely. Priya told me that the biggest resistance came from project managers who were used to being the 'hero' who swoops in to fix problems. Pull planning makes problems visible early, so there's less heroism needed — which is good for the project, but it can be an ego adjustment. Luna: That's a really important point. The system only works if people are willing to be transparent about what's not going well. Lucas: And that's exactly what percent plan complete measures. It's not about blaming; it's about learning. One of the trades in Ray's pilot had a percent plan complete of only forty percent at the start. By the end of the project, they were at eighty-five percent — not because they worked harder, but because they identified and removed the obstacles that were slowing them down. Luna: So what's the one thing you'd tell a listener who wants to try this in their own context, whether it's a software project or a marketing campaign? Lucas: Start with a single project, a single meeting, and a single metric. Don't try to overhaul everything at once. Pick one project that's small enough to fail safely. Get the key people in a room with sticky notes. Ask them to commit to one week's worth of tasks. Then measure how many of those commitments they actually complete. That's it. Luna: And if someone says 'we're too busy for that meeting'? That's the pushback I always hear. Lucas: Then you say, 'You're too busy to save fifteen percent of your project time?' It's a hard argument to win, but you have to have the data on your side. And that's why the pilot is so important — you need proof, not theory. Luna: It's funny, you know, we talk about these systems and they sound so simple in hindsight. But the hardest part is always the first step — actually stopping the old way long enough to try something new. Lucas: Speaking of which — I want to take a quick second to say something that's been on my mind. We get a lot of messages from listeners saying these conversations have actually changed how they run their teams or manage their projects. And that's exactly why Luna and I do this show. Luna: Yeah, we love hearing those stories. And we want to keep making the show without any ads or sponsors, because we think that keeps the advice honest and focused. Lucas: So if these episodes have given you a useful idea or even just a new way to think about a problem — a couple of dollars a month is genuinely what keeps these going. You can find us at buymeacoffee dot com slash fexingo. No pressure, ever. But it makes a real difference. Luna: And we mean it — no pressure. We're just grateful you listen. Now, back to construction. Lucas: So we talked about Ray's conversion. But the really interesting thing happened about six months in, when BuildRight started applying pull planning not just to construction, but to their estimating and procurement processes. Luna: Oh, that's clever. Because the same delays that happen on site — materials showing up late, missing information — those start in the office. Lucas: Exactly. They had a project where the structural steel was delayed by three weeks because the estimator had specified a non-standard beam size. That was a twenty-thousand-dollar mistake. Under the old system, nobody would have connected the dots until the steel didn't show up. But with pull planning, the steel fabricator was in the room from week one, and they flagged it immediately. Luna: So you're catching errors before they become crises. That's the real value of cross-functional visibility. Lucas: Right. And that's the thing I want to leave you with today: the biggest operational gains don't come from working faster. They come from working differently — from changing the way you coordinate. A sticky note and a weekly meeting sounds trivial. But in the right hands, it's a system that can cut your overrun rate in half. Luna: And if you're Ray, you end up with a tattoo you actually like. Lucas: Ha. Maybe a sticky note instead. For The Operations Podcast, I'm Lucas. Luna: And I'm Luna. See you next time.