How Big Things Get Done

Authors

Bent Flybjerg

• Oxford Saïd Business School
• Economic geographer
• Focus of practical decision making
• Main areas of focus: philodophy and methodology of the social sciences, pwoer and rationality in decision making, megaproject planning and management

Dan Gardner

• Investigative journalist
• Also wrote Superforecasting

Introduction

• Two universal drivers for project success and failure:
  • psychology
  • power
• Projects that fail tend to “think fast and act slow”, while successful projects tend to “think slow and act fast”

Chapter 1

• There is a blizzard of numbers in a project, and they are difficult to trust because it’s easy for people to cherry-pick or spin numbers to suit their argument
• “Iron Law of Megaprojects”: over budget, over time, under benefits
• Projects often have a “fat-tailed” distribution (as opposed to normal) where the tails have extreme outcomes
  • Costs aren’t just going over, but theres a non-trivial chance that they go disastrously wrong
  • “fat-tailed” distributions are more likely to occur in complex systems
• Projects that fail tend to drag on because the window is open longer for black swan events
• dynamic inter-dependencies among the parts of the system created strong non-linear responses and amplification
• Focus on slow planning and fast action. Planning is a safe harbor, delivery is venturing across the storm-tossed seas
• Failure pattern:
  • Planning is short, rush to break ground on the project, hit problems not covered in planning, enter a “break-fix cycle”, project drags on.

Chapter 2

• “strategic misrepresentation”: distorting the timeline or costs for strategic purposes, e.g. to win a contract

  “With contracts signed, the next step is to get shovels in the ground. Fast. “The idea is to get going,” concluded Willie Brown. “Start digging a hole and make it so big, there’s no alternative to coming up with the money to fill it in.”

Chapter 3

• Start with the why behind the project
• Good planning explores imagines, analyzes, tests, and iterates.
• Develop a clear understanding of what the goal is why
• At amazon they start with the press release and the FAQ document before starting the project

Chapter 4

• Tinker and iterate the way Pixar and Frank Gehry do
• Experi latin verb meaning “to try” “to test” or “to prove”
  • root of the words experiment and experience
• A good plan apples experimentation or experience, a great plan applies both
• iteration frees people to experiment
• iteration allows scrutiny of all aspects of a plan
• “illusion of explanatory depth” - the fallacy that you truly understand complex phenoma more than you really do
• iteration forces you through the “illusion of explanatory depth” fallacy
• planning is cheap
• This is where the author squares his planning approach with Scrum software practices. In a way, the push to get an MVP out to get feedback is a way of planning and iteration.
  • Author points out that this approach really only works for a subset of projects
  • Where a Minimum Viable product is not acceptable, a maximum virtual product is useful (e.g 3-D rendering), build a realistic model

Chapter 5

• There is not a huge advantage to being first.
• phronesis - practical wisdom - comes from Aristotle
• First Mover vs. fast follower
• Using new technology is just as big a mistake as relying on novice operators
  • Technology is “frozen experience”, tried and tested tools are the result of iterations and lessons
  • An experienced leader has “tacit knowledge” - experience that cannot easily be transferred (for example teaching someone to ride a bike)

Chapter 6

• Often a project is determined to be behind schedule because of delivery issues, but another question is was the schedule/estimation/forecast reasonable to begin with?
• Anchor and adjustment effect plays heavily into project forecasting. You need to pick the right anchor, to pick the right anchor use a reference class.
• uniqueness bias on our project. Our effort is so