Marc Andreessen's startup and career advice from his blog archive

Executive overview

Marc Andreessen wrote extensively about startups and careers a decade ago. His posts — collected in a free ebook — lay out a rigorous framework for founders and early-career people. The core argument: market is the only thing that truly matters in a startup, and most career planning is counterproductive.

Get to product-market fit by any means necessary — everything else is secondary.

Why not to start a startup

• Emotional whiplash is constant: euphoria and terror cycle at speed, magnified by stress and sleep deprivation.
• Nothing happens unless you force it — no established systems, rhythms, or momentum exist.
• Hiring is painful; many candidates flinch at the moment of commitment, even for a legend like Jim Clark.
• Hiring executives is a separate, harder challenge than hiring employees.
• Culture can collapse fast: bitterness and cynicism are self-reinforcing.
• Unpredictable external shocks (market crashes, fraud, regulatory action) can kill the company without warning.

The only thing that matters: market

• Product-market fit is the singular goal before everything else.
• Three variables — team, product, market — all vary widely, but market is the strongest predictor of success.
• In a great market, the market pulls a viable product out of the startup; the team and product just need to be adequate.
• In a terrible market, the best team and best product will still fail.
• Andy Rachleff's law: when a great team meets a lousy market, the market wins; when a lousy team meets a great market, the market wins.
• Markets that don't exist don't care how smart you are.
• Divide startup life into BPMF (before product-market fit) and APMF (after). Before PMF: ignore almost everything else and obsess over getting there.

Dealing with big companies: the Moby Dick theory

• Big company behaviour is largely inexplicable from the outside — often from the inside too.
• Decisions involve dozens of executives, each with veto power; IBM's formal "concurrence" process required 50+ sign-offs.
• A big company can study you for months, then partner with you, disappear, then ship a competing product — with no predictable reason.
• Don't build a startup that requires a deal with a big company to succeed; the risk of never closing that deal is too high.
• Never treat a deal as closed until money or ink is in hand.
• Be extremely patient; never assume a big company will do the obvious thing.

Why the initial business plan doesn't matter

• You cannot predict the right product-market combination upfront; the world is too uncertain.
• Microsoft started with programming tools before IBM forced Gates into operating systems. Oracle was a CIA consultancy before pivoting to databases.
• Edison invented the phonograph while trying to build better telegraph equipment — and didn't recognise what he had for months.
• Aggressively seek a large market and iterate toward fit rather than planning in detail upfront.

Career: don't plan, pursue opportunities

• Rule 1: Do not plan your career. Industries, roles, and the world change too fast to predict.
• Rule 2: Focus on pursuing opportunities — both those that appear and those you create.
• Opportunities present unexpectedly, often while you're engaged in something else. They vanish fast if not seized.
• Think of your career as a portfolio of jobs — each with a risk-return profile, not evaluated in isolation.
• Risk tolerance should vary by life stage: high risk when young and unburdened, lower when family obligations increase, reassessed again later.
• Geographic risk is almost always worth taking to reach the heart of your industry.
• Opportunity cost is a hidden risk — doing one thing means forgoing others.

Skills to develop

• Become a double or triple threat: reach top 25% in two or more complementary skills rather than the best in one narrow area.
• Scott Adams: few people can draw well AND write jokes — the combination is what makes Dilbert rare. Combine skills until your mix is unique.
• Capitalism rewards things that are rare and valuable; you create rarity through combination.
• Five high-leverage skills: communication, selling, management, finance, and technical domain knowledge.
• Most engineers underrate communication because they believe quality speaks for itself. It doesn't.
• Learn to sell: the ability to convince people something is in their interest, predictably and repeatably.

Education and industry selection

• Technical degrees teach rigour, difficulty tolerance, and signal seriousness — regardless of whether you use the subject directly.
• Go to the best institution you can access; being a small fish in a big pond maximises exposure to top people and opportunities.
• Pick industries where the founders are still active as CEO, chairman, or board member — a sign the industry is young, vital, and full of opportunity.
• Avoid industries run by second- or third-generation caretaker managers; they tend to be ossified oligopolies.
• If you come from an elite, orchestrated upbringing, deliberately expose yourself to risk and real failure early — the ability to make tough decisions without good information and recover from screwing up is the most valuable skill you can develop.