I spend most of my time trying to work out how to give good advice: I advise hundreds of founders each year; run a team of advisors; and design the process that our founders go through to get advice.

This is what I’ve learned.

Three types of advice

There are three common types of startup advice: diagnostic; instructive; and method.

Diagnostic advice helps you figure out whether you’re on track or not, and instructive advice tells you what to do. These types of advice are relatively easy for smart people to give, so almost all popular startup advice ends up being diagnostic or instructive.

But method advice, while the most difficult to give, is the most valuable. Method advice isn’t really advice at all if advice means giving recommendations. Method advice just aims to transfer the advisor’s thought processes to the founder.

1. Diagnostic advice

Diagnostic advice is advice that tells you if you’re doing it right or not. It’s usually a heuristic which can be applied in lots of scenarios, and has corollary questions with only one right answer.

Some of the most famous startup advice is diagnostic:

Make something people want. Do your users love it?

Make your product 10x better. What’s the magic moment?

Grow at 7% a week. How fast are you growing?

Be contrarian. What do you believe that other people don’t that’s true?

Work on your edge. Where’s the founder idea fit?

These bits of diagnostic advice tell you whether you’re doing it right. They don’t tell you how to do it right if you’re doing it wrong. They’re useful heuristics for investors in particular – they help filter through opportunities quickly.

Diagnostic advice leaves almost no room for the advisor to be wrong: there’s no suggestion. It simply makes observations about characteristics of successful companies. As the advisor, diagnostic advice is really easy to give.

But diagnostic advice can be very frustrating to receive, especially when there’s no clear way of fixing the diagnosis. It’s patronising – who isn’t trying to make something people want? For good entrepreneurs, diagnostic advice can feel like you’re just rubbing it in.

If you were struggling with a difficult engineering problem, you wouldn’t think your PhD Supervisor was very insightful for suggesting you ‘find a solution that fits within the laws of physics’. That is diagnostic advice, but it’s not a strategy, and it’s not particularly interesting.

Why then, would we think that ‘make something people want’, ‘be contrarian’, or ‘work on your edge’ are interesting?

2. Instructive advice

As a founder, you can respond to diagnostic advice in two ways: iteration; and instruction.

Iteration is where you change your approach based on limited data to try to fix the problem. You make a guess about where to go next. Some people may be better at guessing than others but, if it were this easy, then startups wouldn’t be so hard. To improve your odds, you can look for instruction.

Instruction means you get the answer from someone who knows it already. In issues like accounting or law, this is easy. But, empirically, in startups many fundamental problems are unsolved – there would be many more rich entrepreneurs if anyone genuinely knew how to make something people want. Practically no one can reliably start valuable companies.

The lesson here? For founders, there are no instructors.

It’s hard for any instructive advice, where someone tells you what to do, to be right: each company is unique; the world changes quickly; and advisors have limited context. There are only two answers to ‘are you making something people want?’, and a thousand potential answers to ‘how do you make something people want?’. It’s easy to make mistakes.

As with almost any other scientific enquiry: there is, in fact, a solution to a given problem – an optimal answer for a given company. The hard bit is knowing what it is.

It’s tempting to slip into instructive advice, but you must resist. As an advisor, it massages your ego to give instructive advice, especially if people take it. As a founder, it can be a relief to receive instructive advice, especially if it seems correct.

But acting like anyone has the answers denies the complexity of reality. Instructive advisors are false prophets for founders: no one can make hard things easy.

3. Method advice

Method advice means advising on thought processes, rather than recognising problems or offering solutions. The best startup advisors give method advice.

Just like in science, true insight takes the form of methods for how you might find the solution. All methods are imperfect: there are so many variables in business that any method can at best be a set of heuristics.

Your goal as an advisor should be to help founders use different methods to sketch out possible paths, get them to understand why each method may or may not apply, and push them to make a call. As you go through this, you transfer your methods to the people you’re advising.

It’s hard to do, as making your often intuitive reasoning clear and explicit is hard. But it’s the only way that an advisor can improve the quality of the founders they advise. Diagnostic and instructive advice at best improve outcomes, but method advice improves people.

One EF founder said the most valuable thing they learned at EF was to ‘how think like Teik Guan’. Teik Guan didn’t give him instructions, he gave him methods.

Good methods are often wrong, but they are important step towards rigour: good methods lead to predictions you can test; good methods enable you to simplify complex problems; good methods don’t go out of date. This is why books like The Lean Startup, or High Output Management, have become canonical. They offer a framework for action.

As methods becomes integrated into the MBA curriculum, they tend to develop bad reputations within startups. But the discovery and transfer of methods is really important. Diagnoses aren’t strategies, and instructions won’t last: method advice should be our goal.

Finding methods

Over the last few decades, we haven’t gotten much better at starting startups, given how many people have started them. In no other field where there is, in fact, an optimal solution, would we accept this lack of progress. There are two problems: methods are hard to find; and hard to transfer at scale.

There hasn’t really been a good way of finding methods. Investors are incentivised to diagnose better – find heuristics for differentiating good from bad. Advisors are incentivised to give better instructions – finding ways of getting startups the right outcome. Academics are hardly incentivised at all.

Even if they were incentivised, investors, advisors, and academics couldn’t transfer their methods. They don’t scale. It’s too difficult to get enough first hand data but, even if you could, you couldn’t do much with it. Books and the internet help, but you lose the nuance that produces truly high quality advice, and you lose the relationship which makes people follow it.

This is why I think what we’re doing at Entrepreneur First is important. We work with people from before they even have a company at all, right through to exit. This is not diagnosis or instruction. We are creating a method. The relationship we have with EF founders means that we can transfer our method reliably. We live and die by the quality of our method.

If we nail it, our method will mean to business what the creation of life means to biology; a method to create something from nothing; 21st Century Alchemy.

Thanks to the EF team for reading drafts of this. If you want to figure out our method with us, start here.