Why Ideas Are Cheap

An Idea Worth Less Than a Form

Walk into a VC's office with a brilliant idea and ask them to sign an NDA first. Most will tell you to get lost.[[1]](https://paulgraham.com/start.html)

Paul Graham's reason is blunt: "The market price is less than the inconvenience of signing an NDA."[[1]](https://paulgraham.com/start.html)

That sounds harsh until you look at Google. Its plan was "simply to create a search site that didn't suck."[[1]](https://paulgraham.com/start.html)

So if the idea is that cheap, what actually makes a startup work?

Three Things, No Genius

Graham says a successful startup needs three things: good people, something customers actually want, and spending as little money as possible.[[1]](https://paulgraham.com/start.html)

Most startups that fail, fail at one of these.[[1]](https://paulgraham.com/start.html)

He calls all three "hard, but doable." There is no magically difficult step that requires brilliance to solve.[[1]](https://paulgraham.com/start.html)

Why no genius? A startup offers better technology than people have now, and what they have now is often so bad that doing better takes no brilliance.[[1]](https://paulgraham.com/start.html)

Use It as a Checklist

When a startup struggles, ask one question: which of the three is missing?

Restaurants show the customer test. About a quarter fail in their first year, yet Graham cannot name a restaurant with really good food that went out of business.[[1]](https://paulgraham.com/start.html)

Great food wins even when the place is expensive, noisy, dingy, or out of the way.[[1]](https://paulgraham.com/start.html)

Money has a blunt signature. In nearly every startup that fails, the proximate cause is running out of money, even when something deeper is wrong.[[1]](https://paulgraham.com/start.html)

The checklist's job is to spot which one is missing.

The Doesn't-Suck Heuristic

Google's edge came from three improvements, not magic.[[1]](https://paulgraham.com/start.html)

It indexed more of the Web, used links to rank results, and kept pages clean with unintrusive keyword-based ads.[[1]](https://paulgraham.com/start.html)

Graham turns this into a heuristic: look at something people are trying to do, and figure out how to do it in a way that does not suck.[[1]](https://paulgraham.com/start.html)

In 2005 he pointed at dating sites. In his view, they used simple-minded database matching instead of how dating works in real life.[[1]](https://paulgraham.com/start.html)

Ideas Do Not Transfer

Microsoft's original plan was selling programming languages.[[1]](https://paulgraham.com/start.html)

Its later business model came when IBM "dropped it in their lap five years later."[[1]](https://paulgraham.com/start.html)

That plan change was not a failed idea. Plans often change, and the people behind them keep working.

An idea is a starting point for the people who had it. Nobody else can pick it up and keep developing it.[[1]](https://paulgraham.com/start.html)

Hence Graham's line: "Good people can fix bad ideas, but good ideas can't save bad people."[[1]](https://paulgraham.com/start.html)

What VCs Are Really Judging

Picture two founders. Founder A refuses to pitch until investors sign an NDA. Founder B shows a rough version and listens to questions.

Graham's account predicts investors walk away from A.[[1]](https://paulgraham.com/start.html)

The idea alone is worth less than the paperwork. And VCs mainly hear your idea to judge you, not the idea.[[1]](https://paulgraham.com/start.html)

Founder B gives them what they came for: a look at the founder.

Where Your Effort Goes

If ideas are cheap starting points, guarding a plan or waiting for a brilliant secret idea wastes effort.

The effort belongs on the three requirements instead: good people, something customers actually want, and spending as little money as possible.[[1]](https://paulgraham.com/start.html)

Cheap does not mean any idea will do. The product still has to be something customers actually want.

Curious where Graham's thinking went next?

According to Paul Graham, a startup must solve one magically difficult problem that only a genius can solve.

• True
• False

Answer: False

Incorrect. Graham says good people, customer demand, and low spending are all hard but doable; none requires a magical stroke of genius.

Choose from: good people; customer demand; low spending. A startup needs _____ to keep adapting, _____ to give people a reason to buy, and _____ to preserve cash.

Answer: Choose from: good people; customer demand; low spending. A startup needs good people to keep adapting, customer demand to give people a reason to buy, and low spending to preserve cash.

The three requirements are capable people, a product or service customers actually want, and spending as little money as possible.

A startup has talented founders and strong customer interest, but it spends heavily on offices and advertising until its cash runs out. Which checklist requirement was it missing?

• Good people
• Something customers want
• Spending as little money as possible
• A brilliant original idea

Answer: Spending as little money as possible

The startup was missing low spending. Running out of money is the visible failure, and excessive spending is the evidence pointing to that gap.