Big things are hard to do and even harder to do right. Large-scale projects, such as constructing power plants or implementing corporate software, typically exceed initial time and budget expectations. Only 1 out of 200 large projects comes in under time and under budget1. It applies to small projects, too, like house renovations or work projects. Bent Flyvbjerg and Dan Gardner delve into this issue in their book, "How Big Things Get Done."
Why do big projects go so poorly?
Project sponsors - including politicians and executives - often have a keen interest in securing approval for these projects. As a result, they may deliberately underestimate the necessary costs during the approval process. Even if they know it’ll be more expensive, it’s politically unfeasible to tell the truth. They bank on the project getting approved, beginning construction, and reaching a point where everyone is so invested that they have to complete the project, regardless of the escalating cost - a classic example of the sunk cost fallacy.
People are also fairly bad at estimating - they don’t anticipate unintended delays and issues that might come up. They feel their project is truly unique and therefore can’t be compared to anything that came before it.
We deprioritize experience. Experience is crucial; ideally, you want someone who has undertaken similar work before and learned from past mistakes that didn't cost you. For various reasons, we don’t pick the most experienced parties for a project. 2 examples:
Political: California’s High Speed Rail was a first for the US - we don’t really have high speed rail here, like there is in Europe or Japan. When California’s project started, many foreign high speed rail companies set up shop here to try to win the contract or at least part of it. But for political reasons, US-based contractors with no high speed experience were chosen.
Structural: It is almost impossible for the Olympic Games to be hosted well. Since the location changes every four years, no single city has the opportunity to learn from the past and improve future events.
What can we do?
Adopt a “think slow, act fast” mentality. The financial burden of planning is relatively low. At this stage, nothing is committed; there are no proverbial 'shovels in the ground. Once you start acting or building, though, changes become expensive. So take your time planning and be thoughtful about it.
Find experienced people to help you. As mentioned above, you don’t want somebody to do their first high speed rail project or Olympic village on your dime.
Benchmark against similar projects that have been completed. Rather than merely summing up your estimated costs to formulate your project's budget, consider the expenditure of similar past projects that have been completed. Given that those projects are complete, their expenses incorporate the reality of what typically happens and issues that might come up that you might not anticipate.
As an analogy, imagine you go to the gym for the first time and bench press the 45-lb bar for a set of 5. “That was easy,” you tell yourself, “I’ll add 5 lbs when I go back in a couple of days.” You go back and 50 lbs is easy. Now you’re really excited - “if I add 5 pounds a week, I’ll be benching over 300 lbs in a year. Even if I’m conservative and take 10 weeks off, I’ll still be benching 250 lbs. Not too bad.” This is the inside view - building a bottoms up forecast based on the information you have about your project. The outside view would involve studying how long it takes people to bench 300 lbs, what % of people can actually do it, etc. The outside view is more accurate as it represents reality and what really happens…people get injured and take a step back, they stop going to the gym for a month, or reach a point of slower returns and gains as they get stronger.
Take steps to limit your downside. One challenge with many big projects is that not only is there a risk of something going wrong (10% cost overrun), but there is a risk of something going very wrong (1,000% cost overrun). If you think slow and look at base rates, though, you can minimize some of these risks. What “black swan” events did similar projects face and how can you mitigate them?
Modular solutions. Solar power projects tend to be built fairly efficiently with minimal cost over-runs. This is in part because they are modular, where you are putting together lots of identical blocks (solar panels). Modular projects work better as you get lots of experience doing the same thing over and over again - whether building individual solar panels or baking the mini-cakes that make up a wedding cake. (As opposed, to say, nuclear reactor that can only be tested when fully complete). You can experiment and learn rapidly. And you don’t need to wait until the project is completely finished to test it.
Thoughts and reactions
I really enjoyed this book, as I do anything in the category of “how to execute better.”
I do wish there was more of the discussion on the personal level - what makes someone “good at getting stuff done"? There are obvious things, like high conscientiousness on the Big Five, a high tolerance for pain and confrontation, being persistent, but would love a deeper dive on this (and would welcome any book recommendations)!
The part on benchmarking against similar project was interesting. “Top down” thinking is sometimes looked down upon in the business world - some people might see it as lazy and not as “granular” as bottoms-up thinking, but he does raise a good point that benchmarking and precedent projects gives you a good sense of how long something actually took.
I also appreciated the discussion around modularity and trying to break things down into pieces. Even the way I write these pieces is somewhat modular - I know there will be an intro, a summary, and some reflections. I can write and edit each of these independently as opposed to trying to write one long essay and wait until then to edit it.
Limiting your downside is also important. One mistake people make is to assume that only one bad thing will happen at a time. But what really messes up projects is the “perfect” storm of multiple issues happening at once. Anticipating the “worst case scenario” is important. In investing, for example, people will often “stress test” their financial projections by seeing how returns are impacted if the company grows slower than they thought. The stress test might suggest that the investment could still yield satisfactory returns. However, in reality, if a company's growth slows, it's probable that its profitability and market value will decline as well.
Not every project requires this approach. The book’s approach works for anything that’s important, costly, and hard to reverse. But in everyday work, there are plenty of smaller projects that might be time sensitive or have a lower cost of failure. Or situations where doing something is better than doing nothing2. In these cases, you probably don’t need to be as thorough in your planning; you often just need to act. A key skill of being an effective leader is knowing which projects are important and which aren’t (and communicating that to your team).
Related reading:
Working Backwards: Amazon operates in a way that makes big projects much more likely to succeed. There is a focus on “working backwards,” figuring out exactly how the project will be marketed and questions people might have. This helps people spend more time planning and anticipating potential projects as opposed to just starting.
Flying Blind: The 737 MAX Tragedy and the Fall of Boeing: Designing a new airplane is definitely a new project. Part of the reason the 737 MAX had issues is that there were small problems that were ignored or patched over that got compounded over time.
0.5%
Dumb example: if somebody mails you a gift, it’s probably better to respond promptly in some form - text, email, handwritten card - as opposed to trying to craft the perfect response or pick up beautiful stationary and send 3 months later.