When the Plan Lies to You: What #NoEstimates Taught Me About Construction Scheduling

I’ve been in construction long enough to watch a perfectly color-coded Gantt chart crumble on contact with reality. We’ve all been there. The schedule looks beautiful. The baseline is locked. The client is nodding. And then Week Two happens — and you spend the next six months explaining why Week Two happened.

That’s exactly why my conversation with Vasco Duarte hit different.

Vasco is the creator of the #NoEstimates movement, host of the Scrum Master Toolbox podcast (over 10 years, over 900 episodes — the man doesn’t quit), and a former project manager who once managed 500 people across 100 teams on four continents at Nokia. He knows what a broken planning system looks like at scale. And he’s spent years building something better.

This episode — S7 E116 of The EBFC Show — is one of those conversations I’ll keep coming back to. If you work in construction and you’ve ever felt like your scheduling process is more fiction than forecast, you need to hear this one.

Stop Estimating. Start Counting.

The core of #NoEstimates isn’t “don’t plan.” It’s “stop pretending you can predict the future with a number you made up.”

Vasco’s argument is sharp: estimation as we practice it is a ritual that creates false confidence. We spend hours — sometimes days — estimating task durations. We defend those numbers in meetings. We build schedules around them. And then reality shows up.

“Reality is a bitch. And I say that to anyone who believes planning has magic powers — that you’ll get what you want just because you planned it.”

I’ve quoted that line three times since we recorded. Because it’s true, and it’s something most of us in construction won’t say out loud.

The alternative? Throughput forecasting. Instead of estimating how long something will take, you look at how much you’ve actually done over recent periods. You count. You use real data. You build a forecast based on what the team has demonstrated — not what someone in an office hoped they would do.

In construction, we have a version of this. It’s in pull planning. Last Planner System. Percent Plan Complete. We’re tracking what actually gets done versus what was committed. The data is right there. We just don’t always trust it over the Gantt.

The Nokia Story (And Why It Should Haunt Every PM)

Vasco told a story that’s going to live rent-free in my head for a while.

He was working with a major project at Nokia — 500 people, 100 teams, spanning four continents. They were tracking throughput. The numbers were telling a clear story: the project wasn’t going to finish on time. Not even close.

Six months before the deadline, Vasco went to the program manager and said: we need to cut scope. The throughput data is clear.

The program manager did what program managers do. He sent an email to all 100 teams: “Will you deliver on time?” And of course every team said yes — because in projects, nobody wants to be the first to admit they’re late. You wait for someone else to say it first.

The project got canceled. Hundreds of people’s work, gone — because someone couldn’t bring themselves to trust what the throughput numbers were screaming.

“Believing the plan is why plans fail… The only reasonable thing to do is never believe the plan. Or, as in Scarface — never get high on your own supply.”

I’ve seen versions of this in construction. I’ve watched PMs add float to a schedule they know is broken. I’ve watched earned value reports get massaged until they show what the client wants to see.

And I have my own story — I once worked with a company that had 35 full-time EVM staff. Thirty-five people dedicated to earned value management. Before the meeting started, their lead looked at everyone and said, “Let’s just all acknowledge that earned value management is more art than science.”

More art than science. Thirty-five people. That’s not a measurement system. That’s a mythology department.

The Only Honest Lever: Scope

Here’s the part that construction folks need to sit with.

When throughput data tells you a project won’t finish on time, you have limited options. You can’t manufacture more time. You can’t always add people — adding people to a late project often makes it later. Sound familiar? You can’t conjure capacity out of thin air.

What you can do is adjust scope.

Vasco is direct about this: scope management is the only honest lever. It’s the thing project managers resist the most because it feels like failure. But here’s the reframe — it’s the only move that’s actually grounded in reality.

In construction, we call this value engineering. Phasing. Scope reduction. We do it, but we wait until we’re desperate. Throughput forecasting lets you see it coming early enough to make the adjustment on purpose, not in crisis mode.

The Human Element Nobody Talks About

My favorite part of this conversation was when Vasco talked about Scrum Masters — and by extension, anyone playing that facilitation role on a project team.

He referenced that famous Toyota line: “Ford hires people but only needs their hands. We hire the hands and as a bonus, we get a brain.” That applies to construction more than most people want to admit. Every person on your job site is making hundreds of creative decisions every day.

Vasco said the Scrum Master is the lubricant — not the process police. Their job isn’t to enforce a methodology. It’s to help the team function. To remove friction. To be the person who reads the room and adapts in real time.

I shared a story of my own: a PM who was completely overwhelmed. Ready to give up — not just on the job, but on everything. We started meeting every week. I didn’t pile more process on him. I cared for him first and taught him second.

Within a few months, his throughput quadrupled. And the moment I knew it was working? He told me he was going fishing with a friend on a Friday.

That’s what the human element does when it’s working right. It gives people their life back.

Validate It Yourself

Vasco’s parting advice was characteristically practical.

Don’t take his word for it. Don’t take mine either. Pick 10 periods of throughput data from your project. Split them — first five, last five. Look at the average and the range. See if the pattern holds. Then compare that to what your estimation-based schedule predicted.

“Don’t convince anyone. Convince yourself. Look at the data. Validate it. Don’t believe me — validate it for yourself.”

That kind of advice respects your intelligence. It’s not a sales pitch. It’s an invitation.

Go Listen to This One

I could keep going — we talked about forensic schedulers in construction litigation (they’ve never seen a good schedule), the parallel between pull planning in construction and agile planning in software, and why people defend Gantt charts like they’re defending their firstborn child.

But honestly? Just go listen to Episode S7 E116.

This conversation belongs in every construction PM’s toolkit. Whether you’re running a jobsite or managing a portfolio, the question Vasco is really asking is:

Are you making decisions based on what you believe, or what the data shows?

Here’s the thing I keep coming back to: the schedule is not the project. The work is the project. The data coming off the work is the truth. Everything else — the Gantt, the baseline, the earned value report — is a story we’re telling ourselves.

The question is whether that story is based on evidence.

Or whether we’re just getting high on our own supply.

🎙️ Listen to the full episode: No Estimates, Maximum Results with Vasco Duarte — The EBFC Show S7 E116. Available wherever you get your podcasts.