handbooksThe Lean Startup, explained
What counts as a startup, why most of the work in one is waste, and the loop that replaces guessing with evidence.
- A startup is any team building something new under real uncertainty. Company size does not matter.
- The waste is not slow coding. It is building the wrong thing beautifully.
- The fix is a loop: build a small thing, measure what happens, learn, then decide whether to keep going or change direction.
- Progress is measured in learning, not in features shipped.
What counts as a startup
Ries defines it broadly: a human institution built to create something new under conditions of extreme uncertainty. A two-person app and a new product line inside an eighty-year-old bank both qualify. What matters is the uncertainty, not the funding or the office.
The waste this targets
Most startup effort goes into building things nobody wants. The code works, the deadline is met, the feature ships, and it changes nothing.
Traditional planning cannot catch this, because a plan assumes you already know the customer and the problem. Under real uncertainty you do not. So the plan produces confident, detailed, wrong work.
The loop
The loop is the whole method. Everything else in this handbook is a tool for running it faster or reading its output honestly.
Two things people get wrong about it:
- You plan it backwards. Running it forwards from "what should we build" is how you end up with a feature and no answer.
- Speed means the whole loop. A team that ships in a day and takes four months to read the result is slow.
What this handbook covers
| Section | What it answers |
|---|---|
| The five principles | Why this works, and who it is for |
| Build, measure, learn | How to run one cycle |
| The MVP | The smallest thing that answers the question |
| Metrics that are not lies | Vanity numbers, the three A's, split tests |
| Innovation accounting | How to show progress before you have revenue |
| Pivot or persevere | The meeting, and the ten kinds of pivot |
| Engines of growth | Sticky, viral, paid, and which one you are running |
| Finding root causes | Five whys, and when to use a fishbone instead |
| Inside an existing company | Running this where there is already a business to protect |
Stop asking "can we build this?" Almost anything can be built. Ask "should we build this, and will anyone use it?", then go find out cheaply.