1. Read everything. If you want to do research, read every single paper on the topic (it helps to learn how to conduct a systematic literature review - but keep in mind that generalisations that can be made from any one study are usually very limited). If you want to influence policy, understand all legislation and regulation in immense detail and understand where the political power is. In either case, get as much data as you can on the problem you’re trying to solve.
  2. Speak to everyone who is doing similar work. There are usually projects and documents you wouldn’t otherwise be aware of. Many people are happy to offer advice or help, especially via email. Defer to experts unless you have a good grasp on a topic yourself.
  3. Generate as many ideas as you can. Most ideas suck (“90% of everything is crap”). If you are attached to an idea, you do not have enough ideas. 100 is a good number to aim for, and 300 is better.
  4. For your idea, draw a detailed theory of change and explicitly list your assumptions. Now, you know what hypotheses you need to try to falsify.
  5. Try to shoot your ideas down. Most ideas suck, and the faster you figure out how an idea sucks, the faster you can move on to having impact with the next idea. (Indeed, the ability to shoot your ideas down by using evidence is, for me, the fundamental essence of science.) Shutting a project down is usually more impactful than scaling (unless your idea is actually good!).
  6. As part of the above point, put as much rigour into answering “does this problem actually exist?” as you do into answering “is my solution effective?”.
  7. Reduce the level of confidence you have in most of your beliefs. We are all probably very, very wrong about many, many things. If an idea is sacred in your community (e.g. “it is good to preserve the environment”; “eating vegan is good for animals”; “human extinction would be bad”; “fishing is wrong”), it warrants particularly close scrutiny. Find the skeletons.
  8. Focus on decision-relevant questions. The easiest way to be more efficient is to only answer the most decision-relevant questions and happily ignore the superficial details.
  9. Look for ways to automate or speed up the tasks you spend lots of time on. You’d be surprised at how many technological solutions there are that fix random problems.
  10. In animal advocacy, supply-side policies that aim to reduce production usually suck. There are exceptions, but they’re rare.
  11. Look after your body and your mind.
  12. Don’t be a dick. If you are working against an opposing lobby (e.g. farmers), work hard to put yourself in their shoes. Be especially compassionate for people you have power over (e.g. employees). History is filled with examples of people who wanted to do good but let power get to their head.
  13. Be humble. Altruism (including evidence-based altruism) and animal advocacy have both existed for thousands of years at least (I’m constantly surprised at the parallels between the EA Forum and the Pauline epistles/the Pali cannon/etc etc). We are only here because of the giants who came before us, not to mention the people who donate money to pay our salaries - there is no such thing as individual impact.

Recommended reading: