Everything takes longer than you think

Everything takes longer than I think it will.

I know this. I have lived through it enough times that it should no longer surprise me. And yet I still catch myself making plans as if the clean version is the normal version.

The building project has made this obvious again. A task that sounds simple when described in one sentence turns into a chain of smaller decisions. Someone has to look at it. Someone has to quote it. Materials have to show up. One answer creates three more questions. A thing you thought was settled gets reopened because reality found a detail your plan skipped.

This is not unique to construction.

Software does the same thing. So does hiring. So does writing. So does getting in shape.

The visible task is rarely the whole task. Underneath it is coordination, waiting, rework, weather, and the tiny bits of friction that never make it into the original estimate.

Daniel Kahneman called one version of this the planning fallacy. We tend to make forecasts from the inside view: our specific plan, our motivation, our best-case sequence of events. The better move is the outside view: look at similar projects and ask how long they actually took.

That is easy advice to understand and annoyingly hard advice to live.

A few things help.

Use a reference class. Do not ask, "How long should this take if everything goes smoothly?" Ask, "How long did the last three projects like this actually take?"

Add a margin before you need it. If the first estimate assumes every handoff, quote, delivery, approval, and decision happens cleanly, it is probably not an estimate. It is a wish.

And do a quick pre-mortem. Imagine the project is late, annoying, and over budget. What probably happened? The answer is usually not mysterious. It is waiting, rework, unclear ownership, missing materials, weather, scope creep, or one tiny dependency that holds up everything else.

Part of the problem is that a plan feels like progress. Once the steps are written down, the future starts to look more orderly than it is. But the map is not the work. The work still has to meet the world.

I'm trying to get better at adding a boring margin of safety to my estimates. Not because I want to be pessimistic, but because optimism without a buffer usually turns into stress later.

The point is not to stop planning. The point is to stop pretending the first estimate is reality.

Where are you still planning as if the best-case version is the normal version?
Hit reply and tell me. I read every response.

Have a great week!

Graham

📚 Book Notes

Thinking, Fast and Slow - Daniel Kahneman

Kahneman's book is one of those books that explains an error you were already making before you had a name for it.

The planning fallacy is a perfect example. We look at our own project from the inside and imagine the path if things go mostly right. But the outside view asks a better question: how long did similar things take when other people tried them?

That question is less flattering, but usually more useful.

📖 Article

Hofstadter's law

Hofstadter's law is funny because it is so hard to escape: "It always takes longer than you expect, even when you take into account Hofstadter's law."

I like it because it captures the recursive annoyance of estimation. Even when you know projects run long, you can still underestimate how much you are underestimating.

From the Blog

📝 Best Books for Clear Thinking and Better Decisions
Clear thinking does not make the future clean. It just gives you a better chance of noticing where your assumptions are too neat.

📝 What's actually true about AI data centers?
A useful example of slowing down before deciding what you believe. Separate claims, check the base rates, and resist the first convenient story.

🔗 Things I Found Interesting

The planning fallacy (The Decision Lab)
A plain-English overview of why people underestimate timelines, budgets, and risk even after watching similar projects go sideways.

Reference class forecasting (Conceptually)
The practical antidote: stop relying only on the inside view. Compare your project with what happened to similar projects before.

The Big Dig
A famous example of how complex projects can overrun the original estimate by years and billions of dollars.

Project Hail Mary
A fun fiction version of the same broad theme: the plan keeps changing because reality keeps producing new information.

📣 Quote

"The plan is useless, but planning is indispensable." - Dwight Eisenhower