Here are a number of things I've learned about productivity over the last few years, in no particular order.

  • Structure helps, especially if you're working on your own. Schedule time to work in your calendar. This is a useful commitment device.
  • Find a todo system that works. It should be easy to add elements. Small notebooks are helpful for many people. I've used Todoist for the past year. You want to be in a position where there's always something to work on next.
  • Lynette Bye introduced me to work cycles. The general idea is: break up your day into work sessions. Before each session, deliberate on what you'd like to do and how you'll start and at the end of each session, reflect on how the session went. Ultraworking has their own set of prompts, I've iterated on my own. I break up work sessions into 50 minute increments. Before the session I'll set a purpose. After the session, I'll reflect on what went well, what could have been better, and what I learned.
  • Reflecting should be built into your schedule. It's useful for catching what's going well, what could be better, what you're learning, and whether you're working on the right things. The amount of time you spend reflecting should be proportional to the amount of time you're working on. Spend more time reflecting on larger chunks of time.
  • 2 minute rule: if something will take you 2 minutes do it right away. Sometimes you should be focusing. In those cases, it's important to not make yourself available for 2 minute requests from others or from yourself. Make yourself hard to contact, don't let yourself generate tasks on your own from the internet. If you think of a task, write it down for later. This is the most useful upshot from the getting things done system for me (GTD can be useful for manager schedules, but is unnecessarily heavy for other schedules).
  • Treat work spaces as holy spaces. Beds are for sleeping and sex, desks are for work. However, occasionally it can be useful to switch contexts. Working in the same place can be boring. We're tied to environmental contexts as humans. I've found it useful to program in one place and write in another.
  • The MEVID equation is the most useful things I've read about. motivation = expectancy * value / impulsiveness * delay. Expectancy goes up when your belief that you'll succeed goes up. Think of it as optimism. Value goes up when the task becomes more rewarding. Impulsiveness means exactly what it sounds like. Delay refers to how long one waits before receiving the feedback. We like to work on things that we're optimistic that will succeed and rewarding right now. We procrastinate when tasks become unpleasant, unsuccessful, rewards become distance, and there are shiny alternatives. Each variable in the equation is a lever you can pull to become more motivated. The next few points follow from this simple model.
  • If you're having difficulty doing something, raise the stakes of failure. You can do this with tools like Beeminder. You can also also do this with accountability buddies or tools like Focusmate. If you want to do something, but won't pay $1000 if you fail, do you really want to do it?
  • Productivity is autocatlytic. If you succeed at one task, you'll increase expectancy and be more likely to succeed at the next. Take advantage of success spirals. To do this, your goals need to be realistic.
  • Setting goals and doing them is a matter of self-trust. Don't eat into self-trust. In the past, I've found that it's easy to eat into self-trust by having too many projects at once.
  • Streaks are great.
  • Deep work works. Limit impulsiveness by turning off the internet, hiding your phone, and focusing for a few hours. Start by deciding to do this for short increments of time, then increase it.
  • Wearing earplugs can help focus. Even when things aren't noisy.
  • Habits are automatic behaviors that become a part of your identity. Being the kind of person who checks things off of their todo list can be apart of your identity. Become this kind of person by surrounding yourself with people who value this, consuming content that values this, and doing it.
  • Being process oriented matters a lot. You want to be able to measure the quantity and quality of your work. Work cycles help with this.
  • How much time should you spend working? It depends on the field. Anecdotally, I've noticed that the most successful researchers work more than the typical researcher, but not that much more. Diminishing marginal returns potentially kick in faster for research. If you're in business and you want to succeed at the highest levels, you should work a lot. People at the top of the income distribution work the most. Despite this, if you want to work more, I wouldn't push that much on the abstract goal of "working more." In general, I think the meme of glorifying people working a lot is good, but it can be easy to do fake work in order to satisfy that goal. There's more leverage by working better and ensuring that you're working on the right thing. People who work a lot do it because they like it or are neurotic, not because they feel like they should. I've noticed that I can increase my stamina, but I need to do this by increasing how many minutes I work a day, not hours.
  • Working on the wrong things and becoming socially isolated are real problems, more concerning than the nebulous problem of burnout.
  • One way to ensure that you're working on the right things is to derisk. Look for breaking potential projects as fast as possible.
  • Be direct. If you want to achieve a goal, work on that goal not side goals. If your goal is to speak a language fluently, put yourself in situations where you get as close to the result as possible: move, change the language on your laptop, force yourself to use the language at home. You can learn a language from apps and courses, but because they're removed from the goal, it will take longer.
  • Default to moving quickly. This is likely one of the largest mistakes I made earlier on. I have a tendency to think through things slowly and overemphasizing accuracy. Life is short. Feedback is valuable. If you have a strong error correction system, making more inaccurate decisions can be better than making fewer accurate ones.
  • If something isn't working even though you've tried to fix it, drop it. Accountability buddies are like this for me. Don't just do something because it should work.
  • Relatedly, focus on shipping and publishing. The risk of this approach is that you may be limited by success. This happens with content creators who are captured by markets that don't let the creator create something really great. In general, it seems to me that people who are working without sharing their work face the higher risk of spinning their wheels.
  • The most successful people have an inversion of attitudes: what others see as unpleasant they see as fun. Successful athletes like getting up at 4am. I think of this as the edge of discomfort. If other's don't do X because it makes them uncomfortable, you can do good by learning to like X. Working more is an example of this.
  • Use version control whenever possible.
  • If you set a goal or new habit, ensure that you can measure success. This is obvious, but we often get it wrong. Keep goals challenging, specific and immediate.
  • If you're putting something off, set a timer for finishing it and do it then.