Paul Graham's core ideas on wealth, design, and making great things

Executive overview

Wealth is not money — it is whatever people want. Creating wealth means making things people want, and technology multiplies how much one person can create. The fastest path is a small team with measurement, leverage, and relentless standards.

The founders who win are those who can make genuinely great things and keep making them better — relentlessness wins because unseen details accumulate into something stunning.

Wealth, startups, and leverage

• Wealth is whatever people want; money is just a way of moving it
• A great programmer can create millions in value; a mediocre one generates zero or negative wealth
• Large companies can't reward 10x effort — they lack measurement and leverage
• Startups fix this: small groups are measurable, and you sell directly to customers
• The real point of a startup is 10 people like you, all wanting to work harder and earn more
• Success or failure depends on the first 10 employees — recruiting is the founder's most important job
• Keep the company small in headcount, not in revenue

Technology as leverage

• Technology is technique — a new way of doing things, multiplied by everyone who uses it
• Sam Walton got rich not by being a retailer but by designing a new kind of store
• The lever of technology keeps growing: one person in a bedroom can now affect millions
• Variation in individual productivity will keep increasing; master the new tools or get left behind
• Small companies win on great products; big companies win by sucking less than other big companies

Run upstairs: use difficulty as a filter

• Seek hard problems deliberately — difficulty deters competition
• At ViaWeb: when two features were equally valuable, always build the harder one
• Hard problems are barriers to entry; VCs call it a moat
• In business, nothing is more valuable than a technical advantage competitors don't understand
• Keep that advantage quiet — give competitors as little information as possible

Hackers and painters: learning by making

• Hackers and painters are both makers trying to make good things
• Both learn by doing, not by planning on paper first — sketch, then refine
• Each painting (or product) builds on lessons from the previous one
• Copying the masters is how painters, writers, and programmers all improve
• Learn from examples across domains: a biography of an entrepreneur is a museum of techniques

Relentlessness wins

• A great product has to be better than it has to be
• Leonardo painted every leaf on the juniper bush no one would look at closely
• Unseen details combine in aggregate into something that arrests attention
• Good software has beautiful parts no one is supposed to see
• Fanatical devotion to quality is what separates great products from adequate ones

Collaboration and ownership

• Never have two people working on the same part simultaneously
• Divide projects into sharply defined modules, each with a single definitive owner
• Interfaces between modules should be as carefully designed as programming languages
• Looking at things from the user's point of view is practically the secret of success

Principles of good design

• Simple: when forced to be simple, you must face the real problem and deliver substance
• Timeless: if something is ugly, it can't be the best solution — a better one exists
• Hard: people who do great work all worked very hard; if you're not, you're wasting time
• Looks easy: the conversational ease of good writing comes only on the eighth rewrite — everything is reps
• Resembles nature: nature has had a long time to work on the problem
• Redesign: experts plan for plans to change; mistakes are natural, make them easy to fix
• Can copy: the greatest masters just want the right answer and take from anyone without losing their vision
• Happens in clusters: nothing is more powerful than a community of talented people working on related problems

How to start and how to build

• Start with something clean and simple you would want to use yourself
• The standard is what the product could be, not what competitors happen to have
• Use your own product constantly
• Two things in business: build something users love and make more than you spend
• The cheaper your company is to run, the harder it is to kill
• Great work starts with seeing something ugly and thinking: I could do better than that
• The recipe: very exacting taste, plus the ability to gratify it