You’ve seen the headlines. A new high-speed rail line is five years behind schedule and $4 billion over budget. A massive software rollout for a national bank crashes on day one, locking out millions of customers. A bridge meant to revitalize a city takes a decade longer to build than anyone promised. We’ve almost come to expect it. We joke about "government efficiency" or "corporate bloat," but the reality is much more systemic—and honestly, kind of terrifying if you’re the one holding the checkbook.
Most people think big projects fail because of bad luck. They blame the weather, the economy, or some "unforeseen" technical glitch. But they're wrong. According to Bent Flyvbjerg, an Oxford professor who has spent decades studying over 16,000 "megaprojects" across 136 countries, only 0.5% of large-scale projects are completed on time, on budget, and with the promised benefits. That is a staggering failure rate.
Basically, if you are starting something massive, the math says you are almost guaranteed to fail.
So, how do the 0.5% succeed? Understanding how big things get done isn't about working harder or having a bigger Gantt chart. It’s about fighting against "the iron law of megaprojects," which states that they are over budget, over time, over and over again. To win, you have to flip the traditional script on its head. You have to think slow to act fast.
The Planning Fallacy and the "Lego" Secret
We have a psychological itch to start. When a CEO or a politician announces a grand vision, they want to see dirt moving or code being written immediately. This is "strategic misrepresentation." Basically, people lie about how cheap and easy a project will be just to get it approved, knowing that once it’s started, it’s too big to cancel.
But the masters of how big things get done do the opposite. They spend an agonizingly long time in the "quiet" phase. Take the Empire State Building. It remains a marvel of efficiency, built in just 410 days during the Great Depression. How? They didn't just "start." They planned the logistics so precisely that steel beams arrived at the site marked for specific floors and were swung into place within eighty hours of leaving the furnace in Pittsburgh. It was a factory in the sky.
This leads to a concept Flyvbjerg calls Modularity.
If you want to build something huge, don't build one giant, unique thing. Build a lot of small, identical things and click them together. Think of Legos. A solar farm is highly modular; it’s just thousands of the same panels. A wind farm is just a collection of turbines. These projects stay on budget because you learn how to install the tenth panel faster than the first. You get a "learning curve."
Compare that to a nuclear power plant. Each one is a "bespoke" architectural nightmare. Every pipe is unique. Every weld is a custom job. There is no learning curve because you aren't repeating a simple action. You’re solving a new puzzle every single day. That’s why nuclear projects almost always explode—financially speaking.
Why Your "Unique" Project is Actually a Statistic
One of the biggest hurdles in how big things get done is what psychologists call "inside view" thinking. You think your project is special. You think your team is smarter than the guys who failed last year. You focus on the specific details of your plan.
Nobel laureate Daniel Kahneman suggests taking the "outside view" instead. Forget your plan for a second. Look at a hundred similar projects done by other people. If they all went 40% over budget, yours probably will too, regardless of how "innovative" your new software is. This is called Reference Class Forecasting.
I once saw a project lead insist that their data migration would take three months because they had a "new proprietary tool." Every other similar migration in that company’s history had taken nine months. Guess what? It took eleven. They were blinded by their own optimism.
- The "Inside View": Focuses on the specific case, looks at internal data, and assumes things will go as planned.
- The "Outside View": Treats the project as one instance in a larger population. It asks: "What usually happens when people do this?"
If you want to actually deliver, you have to embrace the boring statistics. You have to assume you aren't the exception to the rule.
The Curse of the "Black Swan"
Nassim Taleb popularized the idea of Black Swans—rare, high-impact events that are impossible to predict. In massive projects, the longer a project takes, the more "window of opportunity" you give for a Black Swan to swoop in and ruin your life.
Think about it. If a project takes ten years, you are practically guaranteed to hit a recession, a global pandemic, a change in government, or a massive shift in technology that makes your project obsolete before it’s finished. This is why how big things get done effectively usually involves shrinking the "delivery window."
Speed isn't about rushing the work; it’s about shortening the time the project is exposed to the chaos of the world.
The Pixar method is a great example of this. They don't just start animating. They create "story reels"—rough sketches with temporary voices. They watch them, realize they're terrible, and throw them away. They do this eight or nine times. They "fail" fast and cheap in the planning phase so that when they finally start the expensive "doing" phase (the actual animation), they know exactly what they are building. They eliminate the "unknowns" before the clock—and the budget—starts ticking.
Experience Is Not What You Think
We often hire "experts" based on their years in the field. But in the world of megaprojects, years of experience can be a trap if those years were spent doing things the wrong way.
The real experts in how big things get done are those who have a track record of "right-to-left" thinking. They start with the finished goal and work backward to the present. They don't ask "What do we do first?" They ask "What needs to be true just before we finish?"
Honest talk: Most project managers are just glorified secretaries tracking delays. They aren't actually managing the risk; they're just documenting the disaster. A true expert identifies the "fat-tailed" risks—those tiny probabilities that could cause total ruin—and spends 80% of their energy mitigating those specifically, rather than obsessing over the color of the office carpet.
👉 See also: Why the 5 languages of love in the workplace are actually about professional appreciation
Real-World Actionable Steps to Scale Up
If you're looking to apply the science of how big things get done to your own work, stop looking for more "efficient" tools and start looking at your fundamental approach. It isn't about the software you use; it's about the philosophy you apply.
1. Find your Lego. Look at your massive goal. Can you break it into small, repeatable units? If you’re writing a book, don't "write a book." Build a system to produce 500-word modules. If you’re launching a business, find the one "unit" of value you can repeat a thousand times. If it can't be modularized, it’s high-risk. Period.
2. Kill the "One-Off" mentality. The moment you say "this has never been done before," you should feel a cold chill down your spine. Innovation is great, but "bespoke" is the enemy of "finished." Use proven tech, proven methods, and proven teams for 90% of the project. Save your "innovation" for the 10% that actually matters.
3. Hire a "Master Builder." Don't just hire a manager. Hire someone who has done this exact thing successfully before. Not someone who has done "something like it." Someone who has seen the specific pitfalls of your specific niche. Experience in building hotels does not translate to building hospitals. The systems are different. The risks are different.
4. Budget for the "Outside View." When you get your final budget estimate, find out what the average cost overrun is for your industry. Add that to your budget. If the project no longer makes financial sense with that "realistic" number, don't do the project. It’s better to kill a bad idea in the boardroom than to let it bleed you dry in the field.
5. Beware the "Beauty Contest." Architects and designers love "monuments." They want to win awards. You want a functional project. Be wary of any design element that adds complexity without adding direct, measurable value. Simplicity is the ultimate hedge against failure.
Understanding how big things get done requires a certain level of humility. You have to admit that you are probably too optimistic, that your project isn't as unique as you think, and that the world is a chaotic place waiting to knock you off course.
The people who build the world aren't the dreamers who ignore the risks; they are the realists who plan for them so thoroughly that the "doing" part looks easy. It’s not magic. It’s just very, very disciplined thinking. Stop rushing into the "doing" and start obsessing over the "planning." Your future self (and your bank account) will thank you.
Focus on the modular. Respect the data. Shorten the window. That is how you win.
Next Steps for Implementation:
- Audit your current project for "bespoke" elements: Identify three areas where you are trying to be "unique" and ask if a standard, proven solution would work instead.
- Conduct a "Pre-Mortem": Imagine it is one year from today and the project has failed miserably. Write down exactly what caused that failure. Now, change your plan to prevent those specific things from happening.
- Consult the Reference Class: Search for three similar projects completed in the last five years. Find their initial budget versus their final cost. Use that ratio to adjust your own expectations.