This guide is based on a presentation I gave during the EAGxVirtual 2020 workshop "How to Run an Independent Project". I've expanded on the presentation to give some personal examples and added some additional details from other sources. Thanks to many many people for valuable feedback, including Michael, Linda, Jaimie, Tessa and Kirsten.
Most of my experience comes from running meta projects. I'm keen to hear suggestions in the comments if there are important differences by cause area.
Is this guide useful for you?
Yes, if you are thinking about running any kind of independent project, that isn’t part of a full-time job. Such projects include: writing a paper or in-depth blog post (or series of posts), creating a tool, website or product, running a workshop or advocating for a cause area. It could even be a personal project, such as researching a particular career path for yourself.
No, if your project requires funding and you are looking for guidance on that topic. It mostly provides advice for people running side projects that don't need funding, at least in the early stages. Suggestions in this guide may help strengthen the quality and/or chances of success of your application for funding. If your project more developed, you might want to look at this Project Management 101 Workshop by Habiba Islam instead.
- Independent projects could be Task Y. Read More.
- Have an open & experimental orientation before starting a project to improve your chances of success. Read more.
- Figure out what makes you motivated, and prioritise it and your morale, wellbeing and happiness. Read more.
- Think carefully about what your goals are before starting a project, and consider whether a project is the best way to achieve them. Read More.
- Take time to choose the right project, accounting for your goals, time, resources and motivation. Read More.
- Create a project proposal and make sure you’ve done your homework, reached out to community members and tried to mitigate possible risks or fall into failure modes. Read More.
- Be open to feedback, and share your progress with the EA community regularly. Read More.
- Share your failures and successes with the community. Read More.
- The google docs version of this guide, the exercises I created and the EAGxVirtual presentation, can be found here.
Part 1: Independent projects as Task Y
There has been ongoing discussion in the EA Community about whether something like a Task Y exists which has the following properties:
Task Y is something that can be performed usefully by people who are not currently able to choose their career path entirely based on EA concerns
Task Y doesn't become much less effective the more people who are doing it
The positive effects of Task Y are obvious to the person doing the task
Task Y is clearly effective
(Edited to rearrange the order)
I don’t think there’s going to be a one-size-fits-all Task Y, because EA is too broad: there are too many cause areas, career paths & goals, and too many potentially promising projects (which is a great problem to have, all things considered).
I think we need to create our own Task Y’s, and for some people small and independent projects could be an excellent way to do this.
Let's do a quick check: Independent projects easily satisfy the first three conditions: they can be done on a part-time basis, there are many possible projects that could be done (many of which haven't even been identified yet), and the positive effects are obvious to the person doing the task (or they wouldn't be motivated do it). Further, as I outline in Step 1 of the Guide., there are many individual benefits from doing an independent project.
The fourth condition is probably the most difficult criteria to meet: are independent projects clearly effective? I think they can be, if we take an exploration orientation. Currently, EA is constrained in many ways which limits our exploration capability.
Thus, if we see independent projects as experiments, then the information we gain as a community, rather than the expected value of any one project, could be extremely valuable and worth the cost. There are many examples of some inspiring projects & research that have been done by EA community members, including (but not by any means limited to):
- Meta initiatives like EA Giving Tuesday, Berkeley REACH, the Technical AI Safety Unconferences and WANBAM. These have been funded to different capacities, but are still mostly volunteer efforts.
- Effective Thesis, Rethink Priorities, Happier Lives Institute, Let’s Fund all started out as smaller or part-time efforts and eventually became full-time projects.
- Loads of excellent forum posts and articles, including many of the winners of the Forum Prize and:
- Guides on various topics, like most of the EA Hub’s group building resources, which have been compiled from various group organisers over the years
- In-depth and well-thought out critiques of popular EA concepts or blindspots
- Series of posts or regular posts, like resource compilations by Michael Aird, or the annual AI Alignment & Literature Review by Larks and donation writeups by various community members.
- Posts that describe different theories, approaches or models really well (and often briefly!) so that they can be easily shared by others, such as this collection of posts and talks on External Movement Building.
- Aaron Gertler walks through some other examples here.
Part 2: The Step-by-Step Guide
Step 0: Make sure you have the right perspective
Keep an open mind, keep experimenting & be willing to fail fast
Don’t get locked into or attached to your concrete-level ideas, especially before you’ve begun. As long as your goal is clear, you should stay open to different ways of achieving it. It may be difficult to disentangle the goal from the idea, which is why spending time on choosing a project, and developing a project proposal is really important.
Keep in mind that successful EA projects have been around for many years, and likely have more funding, time and expertise than your project will. Take an experimental approach to projects, and try thinking about what you could learn if your project fails.
Be willing to abandon a project if you could use your time better
Investing a lot of time and effort can make a project more difficult to abandon, but it's important to be able to do this earlier rather than later. Abandoning a project does not mean giving up after one failure or misstep - it means that you are willing, after enough evidence, to update your evaluation of the project as either infeasible (if your project has repeatedly failed) or outside your current capabilities (if you need to put more resources than you can afford to make the project a success).
Some suggestions on how to give you the freedom to move on to other projects:
- Coming up with a solid project proposal from the start, with clearly defined success criteria, milestones - especially MVPs (Minimum Viable Product) - and scope definition. See Step 4 for more.
- If you haven’t reached or didn’t plan for a milestone, you could still try to salvage your work and make the most of the work you’ve done by sharing what your plan was and why you don’t think it’s the best thing to be doing.
- Get some distance from the project so you can evaluate it more honestly later on.
Experiment with what makes you consistently motivated, and then use it to make progress
Projects are marathons, not sprints. If your project will take more than a few weeks of work, aim to make consistent progress and stay motivated to work on it. A distinction I’ll make here: Motivation isn’t just productivity, it’s also about all the fuzzy stuff - like the things that make you excited, energetic & happy.
You will know what motivates you to do work & make progress. Some people are motivated when framing work as something fun or exciting, while for others it may help to impose stricter guidelines and commitment devices. I’ll sketch a few different motivational models here, it’s likely that most people will be a mix of these, or not even fit into these categories (I’d be interested to hear about this in the comments!)
Intrinsic motivation (finding the project itself interesting, engaging or fun)
There is a lot of uncertainty associated with projects which might make it longer to maintain motivation over a long period of time. If you're just starting out, don't force the project or make it a chore. If over-planning stresses you out, don't do it. If you're exhausted after a hard day of work, go to bed or do something relaxing. You can also take steps to avoid projects dragging on, or to stop yourself getting bogged down by tiny, unimportant details.
It might make sense to take this approach if:
- You have a lot of other commitments such as a full-time job, which take a lot of your time and energy. I am in this bucket, and my general strategy is: if I don’t feel like working on a project, I don’t. Instead I take a break, or work on something else, and come back to the project later on. Although a little haphazard in the short-term, I think it has prevented me feeling burnt out.
- You feel at risk of burnout or are already associating negative emotions with EA or EA-related work (for example, you identify with scrupulosity, and related ideas expressed in the Replacing Guilt series).
Internalising EA values (seeing EA has an obligation; feeling excited, proud of fulfilled by doing impactful things, and adrift when not)
For some people, being more disciplined about EA work is good because they want or need the extra push to do EA-related work. Perhaps doing something EA-related might actually make you feel happier or more energised even if you’re tired.
It might make sense to take this approach if:
- You don’t have a very busy schedule or pre-existing negative feelings
- You know from past experience that this method works for you
- You are fairly confident that making more time for EA-related work will make you happier.
Big picture motivation (your work feels purposeful because its aligned to your big picture goals, like putting the pieces of a puzzle together)
I’m able to maintain motivation and interest in my projects over a long time because they are all feeding into my broader vision of meta work - I’m able to continuously develop my understanding of movement building by working on all kinds of different projects, and use that knowledge to make each project better.
If this motivates you, then regularly remind yourself why your project is important. It can be helpful to take a step back from the tasks at hand and think about how the project you’re working on will help you achieve your long-term goals.
External motivation (getting support and/or being held accountable)
Some people might require external motivation to stay motivated. You can get support and encouragement from people in your social network, like cheerleaders or collaborators. You might also need accountability mechanisms, which could either be self-imposed (e.g. financial penalties) or externally enforced by other people, such as an accountability partner or a stakeholder.
Step 1: To project or not to project?
It may be that a project is not the best way to achieve your goals, and realising this before you start could save you a lot of time. If you’re engaged enough to want to start a project, then there are likely several other things you could be doing that are also worthwhile and will help you have impact over the long run. The following are some quick checks (far from exhaustive) that could indicate whether it’s worth starting a project right now.
If you don’t have a good sense of what your goals are at all, you might want to spend some time figuring them out first, and being as specific as possible.
It may not be worth starting a project if:
You have more pressing tasks which would improve your ability to have impact over the longer term.
Some examples include mental or physical health concerns, general productivity issues or being at risk of burnout. Addressing these issues are important in and of themselves (you do matter!), but these issues will likely be reducing your current or future impact as well. Addressing them could include spending time away from work (EA or otherwise) and with friends, family or hobbies, which would improve your overall health and wellbeing, and reduce your chances of burnout in the long-run. See some detailed examples here.
You are at an early stage in your career and are able to explore other opportunities which would be of more use to you.
Instead of doing particular projects, you could do other activities which help you build credentials, skills or networks in relevant career paths (especially outside of EA) that could be more helpful to achieving your long-term career goals.
Somebody else is working in this space, and you could help them instead.
An anecdote to illustrate the point: Last year, I spent several months thinking about growing EA in developing countries. I wrote various (unpublished) ideas, which were definitely useful to develop my own theory of change but they lacked on-the-ground knowledge of the EA Asia landscape. I'd reached out to EA Asia community builders to share those ideas, and they connected me to Jah Ying Chung, who was doing scoping research for EA Asia at FHI. I ended up helping her with a framework for evaluating the potential of EA in emerging regions which was a really interesting and valuable project, but also got me to update my theory of change in many ways.
It may be worth starting a project if:
You want to improve your long-term ability to have a positive impact and other options aren’t open to you.
A project could help you gain relevant career capital such as upskilling yourself, credentialing or gaining relevant experience. It could also help you test your personal fit, aptitude or preferences for a specific role that you can’t get any other way (e.g. by talking to people or reading). Projects are a great way to try new things.
You have an untested model of the world which would change your actions if you were proven right or wrong about it.
You could test the feasibility of a potentially promising idea, as long as you are open to the possibility of failure and challenges.
The only way to get a good model of the world inside your head is to bump into the world, to let the light and sound impinge upon your eyes and ears, and let the world carve the details into your world-model. Similarly, the only method I know of for finding actual good plans is to take a bad plan and slam it into the world, to let evidence and the feedback impinge upon your strategy, and let the world tell you where the better ideas are. - Nate Soares
You may also be in a failure mode of inaction, where you want to do something but aren’t yet. It could be that you aren't getting sufficient data to make you convinced to do the thing which could length decision-making timelines. It may be possible to do more action-oriented thinking, such as creating a project proposal to properly evaluate your ideas.
So, even if your project is a failure, you can still gain valuable information from it. Your project will likely fail, or fail to have an impact. It’s hard to guarantee impact, even with dedicated time, resources and mentorship that fully-fledged organisations have. This isn’t a reason not to do projects, but instead to be more realistic about what projects you choose and align them with your goals. If you do want to guaranteed impact, choosing projects that are tried and tested (for example, starting a local group) would make more sense than something speculative and uncertain.
You are a “doer” and need project work to stay engaged and learn
For some people discussion is great, but only the first step towards deeper engagement. If you find learning easier and more motivating when it is gamified and there is a concrete objectives, then it might be worth doing a project. For example, community building is a great way to gain an in-depth knowledge of different cause areas, career paths and EA organisations.
Step 2: Generate many project ideas
Step 2a: If you don’t have a specific project idea yet, work backwards from your goals
It may be the case that you don't have a specific project idea in mind: do not fret! This is a great place to be. Just put on your scientist cap and think about your theory of change of the world, and then figure out ways to test it. (Don't worry, this is not as intimidating as it sounds!)
Start with your end-goal or values (e.g. increase donations going to high impact charities) and work backwards from there to brainstorm all the possible ways you might achieve your goals (e.g. coordinate EAs around Giving Tuesday). The last section of this post has a worked out example. To actually find the specific gaps, you can look at collections of project ideas, or simply ask a question.
Once you come up with a few ideas, move on to Step 2b.
Step 2b: If you have a specific project idea, do goal factoring to make sure you've exhaustively brainstormed all possibilities.
Be open to the possibility that independent projects may not be the best way to achieve your goals. Goal Factoring “is a CFAR technique for systematically figuring out all the subgoals and aversions you have around an action, and what to do about them.”
I created this exercise to help evaluate with this process.
Step 3: Do quick feasibility checks for your top project ideas
To evaluate the feasibility of a project quickly, the Murphyjitsu technique is useful because it helps you troubleshoot your project without actually investing time into it.
The exercise I made has a section at the end that walks you through the technique.
Spend some time evaluating the risks of your project ideas.
I recommend watching How to Avoid Accidentally Having Negative Impact with your Project and/or reading 80,000 Hours' Accidental Harm post on this topic for a list of risks to keep in mind as you are doing feasibility checks.
Step 4: Create a project proposal
Even for small projects, it helps to take at least 15-30 minutes and write up a project proposal. A project proposal has many benefits:
- It clarifies your own thinking and theory of change. In the early stages, it’s likely your ideas would not be very focused. Putting ideas to paper helps to clarify and concretise them. This also helps you get more and better feedback.
- You demonstrate to others that you have thought about your projects’ risks and show a basic level of commitment, making it more worth their time to give you feedback.
- You don’t spend too long thinking about an idea without answering some crucial questions.
- It makes funding applications quicker because you don’t have to start grant applications from scratch.
I made a template for the project proposal. Create project proposals that makes sense for your project - a standard research proposal might be better for a research project.
Have a clear vision of your ideal outcomes success/fail criteria
Think about how you will define success and failure, and how you will measure the outcome of your project. This will prevent you from the slippery slope of unconsciously (or consciously) moving the goalposts.
Define smaller project milestones like an MVP (Minimum Viable Product) from the start
- This can prevent demotivation from projects dragging on, and let you get feedback quicker, which you can use to help evaluate your project. You could do this by:
- Split up your project into smaller chunks or share drafts/sections to a smaller group of reviewers
- Apply for opportunities like EAG poster presentations, lightning talks and unconferences to share the progress you’ve made so far
- Share your project & lessons learnt during group meetups or coworking sessions
- Smaller milestones, like cheap tests can be very useful to make sure you’re on the right track. Try to make your cheap tests as easy to evaluate as possible. If you can't do this quickly, then these tests aren't necessarily cheap. Here are some examples of cheap tests.
- Smaller milestones also mitigate risks, because you can test your ideas out on a small scale.
Identify and mitigate potential risks (both for the EA community, but also your own personal growth and development)
- A good first step is to do your homework and look at information available online, including past projects, existing research and relevant organisations. However, this isn't enough.
- You need to actively network and talk to the relevant people. Interacting more closely with the EA community, will help you gain important informal knowledge (which is hard to come by in other ways) of the general landscape and ecosystem of EA. It will help you understand both the quality of your own ideas, but also the external situation within the community that might affect the feasibility of your project.
- Talking to people is the most efficient check for evaluating your projects' risks. Be open to criticism and concern over your project: people will only raise such matters if they think it's important. Just because your project has risks doesn't mean you shouldn't do anything at all, it just means you need to rethink the idea itself.
Consider whether your project would benefit from a collaborator.
Having collaborators to share the work and provide support is always nice, but it can be difficult to find someone who has the same time commitment, goals as you and relevant skills. Your project could fail if a collaborator drops out (which is very common for volunteer projects). Having a collaborator might also make you be too ambitious, which could make the project harder to achieve. Even if you think your project could benefit from collaborators, you could start it on the assumption you will work alone, but let the relevant people know what you’re doing, so people can reach out to you proactively to join. Accountability partners and cheerleaders can help you maintain motivation and provide social support.
Step 5: Get feedback on your project proposal
Get the right kind of feedback on your project proposal. This doesn’t always mean posting publicly or spending a lot of time on it. Depending on the scope of your project, getting feedback at this stage might be as simple as sharing it with a few friends. I’ve outlined all the people who might give you valuable feedback. It’ll be up to you to decide what the appropriate level of feedback is. Getting feedback is an iterative process, but make sure you have define an end-point at which you decide to take the next step, to prevent an endless cycle of feedback.
Engaging with stakeholders is important across cause areas, but especially so in meta and community building efforts, since they are smaller. If you’ve done your homework, and are still uncertain on whether to reach out, then just just reach out. The worst case is that they don’t have time and you can just continue your project with feedback from other community members. In the best case scenario, they could connect you with others, provide you valuable feedback or save you a lot of time by identifying difficult-to-spot red flags.
Try to get some good critics: people with an external perspective (i.e. not a collaborator) whose advice you trust, and who understands your goals and project well. Stakeholders can be excellent critics, but their time will likely be limited.
Step 6: Commit to the project
Set up commitment devices and hold yourself accountable
Get a commitment device, such as an accountability partner. You could have multiple commitment devices for different levels of actions - e.g. collaborators or FocusMate for day-to-day tasks, mentors or cheerleaders for larger milestone goals. Some useful tools:
- EA FocusMate group
- EA Complice Room
- The many, many non-EA resources on accountability mechanisms & personal productivity
Create a support system to keep your motivated and energised
Get some cheerleaders: people who are excited and want you to pursue something and will support your work. I think people with whom you have an informal, comfortable and honest relationship are the best cheerleaders, but you know what would motivate you the best.
Get a mentor: Finding a mentor - whether official or unofficial - is not essential, but is extremely beneficial. Mentors can help you develop meta-project skills and give you direct feedback on your projects. Additionally, if you are doing work that people find useful, potential mentors might reach out to you. If you’re a woman or non-binary person, then WANBAM is an excellent place to find mentors.
Communicate your project clearly to the community to improve coordination and find potential collaborators
Ask yourself who, if anyone, might benefit or act differently if they knew your project existed. You don’t need to publicly announce a project if it’s in the extremely early stages and very uncertain. However, you should communicate it to the relevant subgroups and people, and seek feedback where possible, especially if you suspect there is a lot of information that you do not have access to.
You can announce or publicize your project on Facebook or the Forum, (keep in mind these act like news feeds so you may want to also add yourself to more permanent places like relevant directories like the EA Hub, cause area-specific directories, etc).
When communicating your project, it’s useful to clearly frame the project as an experiment (if that’s what it is). You could share your project proposal, or in particular your goals, the experiment parameters, and success/fail criteria. A past project that was framed well is the Grant Writing Experiment run by Charity Science a few years ago.
Step 7: Complete your first milestone and do a quick evaluation
If your milestone has been successfully accomplished, congratulations! Move on to Step 8.
If your milestone is a failure, come up with a hypothesis on what went wrong, test it quickly and keep iterating until you succeed. If this process takes more time than you initially planned for or ends up changing the scope of your project, you may want to reconsider whether the time investment is worth it.
If you decide the project should be abandoned, skip to Step 9.
Step 8: Repeat Step 7 for all milestones until you've completed your MVP
Step 9: Evaluate your project and consider next steps.
Go back to your original project proposal and evaluate your project against your expected outcomes. How successful was your project?
If you think it was a success, then repeat steps 4-8, defining your next MVP (skip step 6 if you have a good commitment system going!)
If you think your time is better spent elsewhere, or the project was a failure (or both), share your failures and lessons learnt so that the broader community can benefit. Some excellent examples of this:
- Celebrating Failed Projects Panel at EAG 2017
- Ineffective entrepreneurship: post-mortem of Hippo, the happiness app that never quite was by Michael Plant
- Many EA organisations have Mistakes pages
So what are you waiting for? Go forth and project!