This section could be called “How to Avoid Failure.” But the truth is, you can’t.

If you’re aiming big, you’ll inevitably end up with a project in your career that never makes it to the finish and sometimes fails to get off the starting line. You may well have many of them. Whatever you do, don’t adjust your sights to avoid the big stumbles. Yes, failure is humiliating, but it’s the price of going big. You must be willing to fail. Otherwise, you get caught in the aspiration trap, lowering your aims in the hopes that no one notices if you fall.

That said, there are ways to keep your big bets on track. On complex issues, there will be times when you do everything right, but people will say no or worse. It’s the nature of this work and the complicated challenges we face.


Failure is never fun, but big failure is just brutal.

Fail fast when you must.


The important thing is to really know who you’re dealing with. Here are a few ways I’ve learned to do so:

  • Stay grounded. Realize that previous successes don’t beget the next big win.
  • Who is the leader? Motivations matter, a lot. And yes, there are people who would rather enhance their own fortunes than do something to benefit millions. Do your due diligence. Understand what parties stand to benefit, and how.
  • Test the trust. Alliances depend on trust  — you need to know if there is enough trust between partners to get to scale. Stress test it.
  • Be willing to fail fast. Every minute spent on a project that has no realistic future is a minute wasted. Even though failure is a bitter pill to swallow, it’s up to you to be okay with accepting that risk and moving forward with the next task.
  • Know who can say no. Who has the power to end your project? Identify all possible perils and develop contingencies to deal with them should they arise.

From the book:

The truth is, you cannot always avoid failure — something I learned when trying to make the Grand Inga Dam in the Democratic Republic of Congo a reality. Really knowing your partners is a necessity and will require you to stay grounded, test trust, fail fast rather than fail slow, and know who can say “no”. I learned that all the hard way with a big public failure on Inga.