Move fast and break things, and count what breaks
The slogan is sound arithmetic for a website that can roll back in minutes. It is not a law of speed.
Move fast and break things. Unless you are breaking stuff, you are not moving fast enough.
Facebook’s engineers work under this motto, and Zuckerberg repeated it to Business Insider’s Henry Blodget in 2009 when asked how the company kept innovating as it grew. The first sentence is the slogan. The second makes the bolder claim, because it offers a test: broken things are evidence of speed, and their absence is evidence of caution.
The reasoning is about cost. At a company that pushes new code to a website every day, most mistakes are cheap. A broken feature can be rolled back in minutes, and the team learns something from the attempt. Slowing every change down to avoid those mistakes costs more than the mistakes do, because the slower company tries fewer things and learns less. Facebook built its release habits around this, with small changes shipped often and engineers expected to own what they put live.
For a young startup the case is stronger still. A company with a few thousand users has little to break and a great deal to learn, and the users it has are mostly early adopters who expect rough edges. The balance shifts as the audience grows, since each mistake reaches more people, and that is roughly when companies begin adding the review steps the motto was written to resist.
The rule holds under the conditions it came from: web software, fast rollback, and users who shrug at a glitch in a free service. It holds much less well where a mistake cannot be undone. A bank that breaks its payment system, or any service that exposes its users’ private data, has done more than move quickly. The damage lands on people who never chose to take the risk. Facebook’s own privacy changes drew heavy criticism this year, a reminder of how the slogan’s arithmetic looks from the user’s side.
The second sentence is also weaker than the first. Breakage is a side effect of speed, and a poor measure of it. A team can break a great deal through carelessness while moving slowly, and a well-run team can move fast and break little. The useful form of the advice is narrower: decide which mistakes are cheap, make those freely, and spend care on the few that are not.