Advice is dangerous. Like a horoscope, advice often sounds well-crafted and relevant, but believing it is rarely productive. With clever turns of phrase and time-honored proverbs, you can turn any position into an elegant, self-evident truth. “One bird in the hand is better than two in the bush” say some, and others respond “never compromise, doubling down is the only way to succeed”. “The customer is always right” but also “sometimes you gotta fire your customer.” Warren Buffett is quoted religiously for his investment advice, and a big reason for that is that he has given so much advice in his career that you can quote him to argue one way, and quote him for the counter-argument too. What is one to believe?

Programmers learn this early on in their careers. On a late night debugging an issue, they’ll Google the error message to find… no results. Every other issue could be solved by copy-pasting from StackOverflow, but not this one. A sinking feeling sets in. You’re on your own now. No advice will help you now, you have to figure it out yourself.

In other, less mechanical domains, such moments are rare. There’s always relevant-seeming advice. Career advice is full of idiosyncratic and contradictory positions, and business advice legendarily so. The contradictory diversity is because when people give advice, it tends to directly reflect their own experiences,1 and most people’s experiences are quite different from one another. Businesses can be startlingly unique in the details, so it’s no surprise that one person will tell you to focus on scalability, and another will tell you to do things that don’t scale.

Because advice is usually a generalization of “this worked for me”, it’s important to see that advice is usually reasoning by analogy, which is dangerous. “If we say that your situation is like mine fifteen years ago, then you should do this…” is the silent preamble that should caveat most advice. It breaks the same way that reasoning by analogy does: surprise surprise, your situations are not actually analogous. They’re different in the details, and the details matter. Most advice is not good or bad, but inapplicable.

But surely there exists useful advice? Yes, there exist playbooks. At scale, collectives of people have figured out roughly how to optimize a process. There’s a playbook for getting into a good college. There’s a playbook for interviewing at Amazon, starting a B2B SaaS company, or becoming a Chess Grandmaster. These playbooks are syntheses of successful approaches, like people figuring out together how to speed-run a video game. However, in the real world, the territory is always changing. These playbooks, even good ones like YC’s Startup School, have expiration dates.

What are you to do if there’s no playbook? Well, you’re on your own. Silicon Valley types would say “think from first principles”, which is generally your best bet, but it can be hard to get the premises right. More importantly, it’s time to trust yourself. If you are in an area with no playbooks, you have graduated to the frontier. You are there because you are good at what you do. If you are good at what you do, then why not double down on your own intuition? Do what you think is best is usually good advice to those who have already shown exceptional capabilities.

The irony that this is an advice post does not escape me. Of course it follows my own experience: in my younger days I spent much time seeking mentorship and advice. The advice I received was generally not very good. I remember running my first prospective angel investment past a number of venture capitalists I respected. They all passed, citing various qualms. The investment seemed like a no-brainer to me, but given the bearishness of these experts, I passed, thinking “surely there’s something I’m not getting here”. In the end, all the investors were wrong, my intuition was right, their advice was just noise and I missed a phenomenal deal. From there on I stopped listening to what others said, and strictly followed my own judgement: if I’m here, I got here because I’m good at what I do, and so I’ll just trust myself.

Life is full of noise. Advice is a combination of people’s genuine opinions and their performative signalling, repeating the tenets of their in-group.2 This form of noise is omnipresent, and it’ll pull you in all directions. You could spend all day listening to two business gurus give perfectly contradictory advice; it all nets out to zero. For the most part, just listening to it will muddy your thinking and lead you through all sorts of mental contortions to conform to the intellectual aesthetics of others, instead of carving the way you believe to be true. There’s a way to cut through the noise: ignore it, trust yourself, research carefully, reason rigorously.



  1. This is excepting an even more dangerous form of advice, the purely theoretical advice given by someone who has zero practical experience with the subject matter. 

  2. Extremely common in business contexts. “Naval says…”, “Charlie Munger says…”, and so on. These are fortune cookies not intended to communicate any meaningful truth about the world, but to signal that you follow the same gurus and belong to the same group.