Jeff Bezos: Amazon’s secret of avoiding project disasters

Did you know that in the early 2000s, the US Air Force, US Army, and US Navy ranked first, second, and third in the number of aircrafts they owned? The whole of Russia ranked fourth.

  • US Air Force: 6000 aircrafts
  • US Army: 4000 aircrafts
  • US Navy: 3700 aircrafts
  • Russia: 3000 aircrafts

So someone at the Pentagon did a common sense analysis: if all the US military branches combined forces and consolidated their development efforts, they would save a lot of money in the future. One interoperable aircraft that all 3 branches could use – which would replace their old ageing aircrafts. That’s how the idea for F-35 was cooked up.

And F-35 is a thing of beauty. Stealthy. Advanced sensors and targeting capabilities. Can be used in reconnaissance missions to ground attacks. It gave complete air superiority.


The F-35 were supposed to be operational by 2010. And estimated to cost $200 billion. But that didn’t happen. They became operational only in 2015-2016. And the estimated cost ballooned to $1.5 trillion!

What went wrong?

Feature creep happened

To keep all 3 branches of the military happy, minor changes were made. Fuselage was lengthened by 5 inches to add advanced avionics. Horizontal stabilizers were moved 2 inches. The top surface was raised by 1 inch to support a weapons bay. Adding advanced fighter aircraft systems added weight. Which made vertical landing impossible.

So then more changes were made to reduce the weight. And every change led to more change requests from one of the military branches.

The reason for feature creep

When project outcomes are not clearly articulated and defined from the beginning, feature creep inevitably happens. Good intentions get undone.

It feels like every new feature request is an improvement. But it increases the complexity, cost, and delays.  

The root cause is ambiguity. The way to fix it is to create clarity from the get go. Define the end before you begin.

How Jeff Bezos creates clarity

Jeff Bezos, the founder of Amazon, has a rule for product development. He realized that after every project was launched, someone from the marketing team wrote a press release and a FAQ. So he made it a company policy: work backwards. Write the press release and the FAQ before starting a project, not after its launch.

  • The press release articulates who the audience is, the problem they are facing, how current solutions fail, how Amazon elegantly solves the problem.
  • The FAQ helps anticipate customer concerns and objections. And tells people how to get started.

This gives people clarity. They know what a done product looks like. They know what a customer would need. Feature creep is minimal.

It’s this working backwards strategy that allowed Amazon to create everything from the Kindle and Amazon AWS under budget.

Action Summary:

  • Start with clarity. Start only once you know what the end product should look like. Then work backwards.
  • Define the problem + solution. Work out the potential objections. And your product becomes sharper. Selling your product becomes easier too.