Building an Impact-focused Community

Acknowledgements: A huge thank you to the Hive team and the many community builders who have shared their wisdom with us over the years. This post is an attempt to synthesize those lessons. Special thanks to Therese Veith, Gergő Gáspár, Sam Chapman, Sarah Tegeler, and John Salter for reviewing this post. All mistakes and oversights are our own.

TL;DR:

This post walks through some of our key insights from building Hive, a global community for farmed animal advocates, and other communities within and outside of the EA space. We walk through a three-phase approach to community building, and extensive notes in the footnotes. In short:

1. Phase 0: Solve a Problem, Don't Just Start a Group. Before you build anything, a specific group of people needs a compelling reason to gather.
2. Phase 1: The Cold Start Problem & Initial Growth. Obsessing over your first users, doing unscalable things to provide value, and being strategic about who you invite in.
3. Phase 2: Scaling, Setting up Systems, and Navigating the Messy Middle. This involves handling growth, building systems, and cultivating a culture that lasts.
4. Phase 3: Maintaining and Supporting the Ecosystem. This involves continuously improving your community, staying hands-on, and seeing your community as infrastructure for the wider movement.

We also cover common pitfalls we've seen, like underestimating the workload and losing focus, as well as topics like handling conflict, preventing burnout, and knowing when to stop or sunset a community.

Why We're Writing This

Lately, we've been getting a lot of questions about how to start and grow communities. This post is our attempt to summarize the lessons we've learned while building Hive.

This isn't a definitive guide, but rather a collection of reflections, tips, and mental models that have helped us. Community building is ultimately personal work, but we hope that by sharing our journey, we can help others navigate theirs. While our experience is rooted in EA communities—particularly farmed animal advocacy—most of these principles apply broadly to mission-driven communities. As is common with such collections of tips, take what you find valuable and leave behind what you don’t!

Phase 0: Solve a Problem, Don't Just Start a Group (Before you launch)

It's tempting to start a community based on a shared identity ("let's gather all the EAs interested in X"). But we've found that the most durable communities combine identity with a concrete problem-solving approach. A shared interest or identity isn't enough to sustain a community over time.

Here are some questions we've found helpful when assessing the need for a new community:

• Where are your people now? If they're all in one place, maybe you just need a channel or a sub-group within that existing community. If they're scattered everywhere, that fragmentation is your opportunity.
• What new value is created by bringing them together? What insights, resources, or connections will they gain counterfactually in your space? If they can get most of this value elsewhere, your community might struggle.
• Why would they choose your space over existing ones? Your unique value proposition should be clear and compelling. For a potential member, the choice should be simple: "If I want to do X, I go to [existing space]. But if I want to do Y, I come here."
• Is this worth the significant time investment? Building a thriving community often requires at least 15-25+ hours per week from at least one dedicated person. Depending on the scale of your community and your vision, this could require vastly more. Do you have the capacity, and is this the highest-impact use of your time? If you're testing whether community building is right for you, you might start with 5-10 hours weekly for a small pilot, but successful communities rarely remain side projects for long.

Phase 1: The Cold Start Problem & Initial Growth (Months 0-3)

The beginning is all about delighting your first users. Forget about big numbers. Your only job is to create an exceptional, high-touch experience for the few people who show up, overcoming a cold start problem.

• Choose the right platform: We chose Slack because it's a high-quality tool and, more importantly, because our target audience was already using it for daily work. Slack can get expensive, and other options like Discord or Facebook Groups all have pros and cons. Ideally, your platform should meet people where they already are and provide the functionalities necessary for the type of interaction you imagine in your community.
• Understand who you (don’t) want in your community: Your early community members set the culture for those who are yet to join. Many communities start off with a facilitated program, but if your prospective community members are not filtered through such a mechanism, you will need to be more strategic and clear.

  • Consider early on: Will your community stay deliberately small (maybe 50-100 members) to maintain intimacy and high trust? Are there social or political risks to consider? Or are you building something that scales up consistently in value as it grows? This decision should inform how selective you are from day one.^[1]

  • Being intentional about diversity and inclusion: Your early members set the culture. If your first 20 members are all of similar backgrounds, ages, or perspectives, you risk creating invisible barriers for others. Actively seek out diverse voices in your founding cohort. Diverse communities tend to solve problems better and feel more welcoming to newcomers.
• Embrace multiple funnels for outreach: Start with several avenues to bring people in. For instance, you can go to relevant conferences and promote in existing, adjacent communities. This gives you good control over who joins your community. But it also ensures that there is unique value in your community because you are reaching people who are evidently fragmented from a variety of places.
• Focus on giving, a lot: In the early days, you are a significant part of the community's value. You make the introductions, you find the resources, you answer the questions. We spent many hours in 1:1 calls with our first members, learning their stories and figuring out how to help. This doesn't scale well directly, but it builds the base you need to scale. Especially in the early days, your goal is to provide immense value before asking for anything in return. People will stick around if they feel like they've found a place where they are seen, valued, and supported.

Phase 2: Scaling, Setting up Systems, and Navigating the Messy Middle (Months 3-12)

This is often the hardest phase: Initial excitement from launch has worn off, but you haven’t yet established stable engagement. A few strategies helped us:

• Create reliable rhythms: Even if it’s just one monthly event or a weekly discussion prompt, consistency helps people know when to show up.
• Celebrate small wins publicly: Highlight connections made, problems solved, or resources shared. This reminds members why they’re here and helps build a stronger pitch to new members/supporters.
• Accept the dip: Engagement will fluctuate. This is normal. Focus on serving the people who do show up, not obsessing over those who don’t. The exception, of course, is that you are just missing your target audience(s).
• Ensure a well-designed onboarding process: As your community scales up, it can quickly become difficult for new joiners to understand how to best leverage your platform - and you won’t be able to have a 1:1 call with everyone. A great onboarding process can ensure that you set people up for success. A simple form + good intro message or email series setup can go a long way.

• Create guidelines and policies: While not strictly needed early on, the earlier you set them up, the better. We set up basic guidelines and policies right away and evolved them over time. We have found that it’s much easier to moderate and keep your community focused if you have written down policies that you can refer people to. Our own community guidelines are constantly evolving and often the result of things going wrong. You can ask existing communities for guidelines and moderation policies so you don't have to start from scratch.^[2]

• Encourage live interaction:. Regular video calls, co-working sessions, or in-person meetups can deepen relationships in ways text channels cannot. 1-1 matching systems, such as Donut intros (we got a free 12-month trial when we started out!), facilitate calls between community members without you needing to do anything. Find the right mix for your community’s needs and geographic distribution. This might shape how you perform outreach, fundraising, source volunteers, and where you hire.
• Handling growth spurts: Occasionally, you’ll experience sudden influxes (think 50+ people joining after a conference, viral post, or new funding). Plan for this.  
  • Pause open invitations if needed.
  • Create a dedicated “welcome” or “introductions” channel where new members can connect.
  • Consider a special onboarding session or welcome call for large groups.
  • Ensure your moderators can handle the increased volume.
• Make sure that people have a way of interacting with each other, not just have 1:1s with the community organizer. For example, a newsletter is not a community because it’s a form of one-to-many communication, even though it unites people who are interested in a particular topic. If they start interacting in the comments of the newsletter with each other, that’s already the beginning of a community. A one-off event is not a community, but it can become a community if the organisers or attendees find a way to consistently keep in touch with each other.
• Get community members involved in operations: You usually can't scale alone, and for many communities, volunteers can be of great support. We set up a volunteer system to help with onboarding, channel leading, and moderation. This reduced our workload and created a deeper sense of ownership in the community.  Volunteers also make great first hires; in fact, almost everyone who works at Hive was a volunteer first.

Phase 3: Maintaining and Supporting the Ecosystem (12+ months)

Note: In practice, the line between "scaling" and "maintaining" is incredibly blurry. You don't just finish scaling and then start maintaining; the work of maintenance actually begins during the scaling phase.

Once you have a core, engaged group, your job changes; you move from manual attraction to intentional nurturing and scaling.

• Give your members multiple ways of giving feedback and iterate based on what they would like to see: You can’t do everything they ask, but you can gradually improve the community by regularly considering their needs. Some ways you can do this:
  • Create a public channel for giving feedback to you.
  • Create a form that people can use to provide feedback anonymously.
  • Use 1:1 conversations like chats, calls, or meetings to ask about the members’ experience and ask what they’d like to see you add or improve. By far, most of the feedback given to us comes from 1:1 conversations, but the public and anonymous channels are important systems to set up for transparency and confidentiality purposes.
• Protecting scope and saying no: As your community grows, you'll get many requests to add features, programs, or expand focus. Most of these requests come from a good place, but saying yes to everything leads to dilution and burnout. Practice saying: "That's a great idea, but it's outside our core focus of solving [specific problem]. Have you considered [alternative resources]?" Your community's value comes from doing specific things well, not from being everything to everyone.
• Stay hands-on: This is crucial and often missed. Community infrastructure is not a "set it and forget it" machine. You still need to be in the channels, making connections, and prompting engagement and support. If you have ever nerded out about community building with us, you may have heard the behavioral model B = MAP (Behavior = Motivation + Ability + Prompt). The platform provides the ability for motivated people to realize impactful opportunities, but you often need to provide an additional prompt. A personalized DM, an introduction to another person, or a forwarded opportunity can make a massive difference, and we have found that relying on the infrastructure can often leave an impact on the table.
• Help shape and support the wider ecosystem: A healthy community sees itself as infrastructure for the wider ecosystem, not as the ecosystem itself. Your goal isn't to keep everyone inside your walls, but to make the entire landscape of organizations and resources easier for your members to navigate and gain context in. This means actively connecting people to other key players, building strong referral systems, and collaborating instead of competing. Success isn't just about what happens in your community; it's about whether your members can easily find what they need to have an impact, wherever that may be.

Knowing When to Stop or Sunset

Not every community should, or will, exist forever. Sometimes, the most impactful decision is recognizing when a community has served its purpose.

• Signs it might be time to sunset:
  • The original problem you solved has been addressed (or is being better addressed elsewhere).
  • Engagement has declined despite multiple revival attempts.
  • Leadership cannot be sustained, and no suitable successors exist.
  • The community has drifted so far from its purpose that it no longer provides unique value.
  • Maintaining it prevents higher-impact work.
• How to sunset gracefully:
  • Give members plenty of warning (3-6 months if possible).
  • Explain honestly why you're making this decision.
  • Help members find alternative communities or resources.
  • Archive valuable content/knowledge in an accessible way.
  • Celebrate what was accomplished together.
  • Consider whether a smaller, lower-maintenance format might work (quarterly newsletter, annual reunion).

Common Pitfalls to Avoid

These are some of the most common mistakes we see, and a few we've made ourselves.

• Underestimating the workload. A great community is rarely a side project. It needs someone who sees it as their "baby" and dedicates significant time to it (we'd say 15-25h/week minimum).
• Overpromising and under-delivering. It's far better to do a few things exceptionally well than many things poorly. Don't announce a mentorship program, a job board, and five weekly events all at once. Start small, deliver consistently, and build from there.
• Slow moderation. An unmoderated or poorly-moderated space can lose its value very quickly. Have clear guidelines and be prepared to enforce them. The emotional difficulty of moderation causes many builders to avoid it, but timely, kind enforcement is essential.
• Losing focus. The value of the community is solving a specific problem. It can be tempting to let the conversation drift everywhere, but that can dilute the core purpose and make it less useful for everyone.
• The post-event hype. Excitement at an event often leads to commitments like "let's have a weekly call!" that fizzle out. People are busy, so before committing, ask: "How will this call actually solve a problem for you?" and "Is it a better use of your time than what you'd otherwise be doing?" Starting with a sense of momentum is great and important, but it is much more difficult to stay in touch once a conference or retreat is over than you’d often expect.  

• Chasing vanity metrics. Your metrics should be aligned with your organizational goals (and the requirements of funders). E.g.: Total member count is likely less important than engagement metrics. A community with 500 members and 200 weekly active users is likely healthier than one with 5,000 members and 100 weekly active users.^[3]

• Ignoring the emotional labor. Community building involves absorbing others' stress, navigating conflicts, and maintaining enthusiasm even when you're exhausted. If you don't acknowledge and plan for this emotional cost, burnout is far more likely.^[4]

Share Your Journey

As is evident by the number of “notes” we felt compelled to add, there is a lot to say about community building. And certainly, there is a lot that doesn’t apply to everyone and a lot that hasn’t been said.

If you’ve built or been a part of a community that you love, I’d be curious to hear:

• What was the specific problem it solved for you?
• What are the small, often invisible things that made it feel like a place worth returning to?
• What is an underrated tip that has made a community work for you?
• How did the community builders handle the hard moments (conflict, transitions, or difficult decisions)?

Appendix: Quick Start Checklist

Right, so, this was a lot. We put together a short, practical roadmap that you might consider adapting as you’re getting started:

Phase 0: Before you launch

• Identify the specific problem your community will solve
• Confirm people are currently scattered/underserved (not already in one place)
• Determine if you have 5-10 hrs/week to dedicate (at least initially) and 15-25+ hours long term

Phase 1: Months 0-3

• Choose a platform that your target audience already uses for lower entry barriers
• Recruit 10-30 founding members through personal outreach
• Have 1:1 conversations with each to understand their needs
• Create basic guidelines and community norms
• Start giving value before asking for anything

Phase 2: Months 3-12

• Develop a simple, scalable onboarding process
• Evolve your guidelines, community norms, and policies
• Create 1-2 regular rhythms (weekly discussion, monthly call, etc.)
• Recruit 2-3 volunteers from engaged members
• Start documenting your processes

Phase 3: Months 12+

• Stay hands-on
• Assess whether growth should continue or be capped
• Implement formal moderation procedures
• Create a sustainability plan (funding, leadership succession)
• Gather systematic feedback from members

1. ^{^}

   A Note on the Right Community Size: While every new member adds value, they may indirectly reduce the average willingness-to-help per community member. That is, as a community gets bigger, the relative value each person can find or contribute tends to decrease. We’ve seen this happen for a few reasons:

   • Overwhelm: As the volume of posts and conversations grows, it becomes harder for anyone to keep up. The signal gets lost in the noise, and individual engagement drops.
   • The Bystander Effect: In a larger group, the sense of personal responsibility to help with a specific request diminishes. It's easy to assume, "Someone else, who is more of an expert, will probably answer this."
   • Loss of Connection: The close-knit, high-trust feeling of a small group can fade, making it harder for people to feel personally connected and invested in each other's success.

   When to slow down or stop growing: Not every community should scale indefinitely. Consider slowing or capping growth when:

   • Your core value proposition depends on intimacy (mentorship circles, support groups).
   • Engagement metrics are declining as you add members.
   • Your moderation capacity is consistently overwhelmed.
   • The community is serving its purpose well at its current size and would do so less if it were larger.

   Now crucially, none of these problems is impossible to solve. Impressive mass movements can scale up to tens, if not hundreds, of thousands of people, many of whom retain a strong sense of agency and willingness to help. The challenge is to scale without losing the density of helpful interactions that made the community valuable in the first place, and to know when to stop/slow down.

2. ^{^}

   A Note on Effective Moderation: Slow or absent moderation is one of the fastest ways to destroy community value. Here’s what we’ve learned:

   • Effective moderation practices:
     • Respond quickly to guideline violations (within 24 hours, when possible): Delayed response signals that rules don’t matter
     • Start with private, kind conversations: Most issues resolve when you DM someone saying, “Hey, I noticed X. Could you help me understand what’s going on?”
     • Document everything: Keep records of moderation decisions for consistency and learning
   • Have a clear escalation path
     • When to remove someone: This can be emotionally and socially difficult, but it is sometimes necessary. Remove someone when:
     • They repeatedly violate guidelines after warnings
     • They create an unsafe environment for others (harassment, discrimination)
     • Their presence is causing valuable members to disengage or leave
   • Handling interpersonal conflicts:
     • Often, members will bring disputes to you privately or anonymously. Your role is usually a mediator, not to judge
     • Help both parties understand each other’s perspectives; facilitate direct (moderated) conversations when appropriate
     • Sometimes, the best resolution is helping people agree to avoid each other within the community space (which could mean the complete removal of one or both)
3. ^{^}

   A Note on Impact Measurement

   It is notoriously difficult to measure impact for meta work in general, and community-building work in particular. The outcomes we hope to achieve are very varied and indirect in nature. We ourselves have previously spoken and written about how we evaluate our community-building impact (and the difficulties in doing so). You'll often find lead metrics such as active members or event attendees, and lagging indicators such as testimonials/stories of people detailing how being part of a community has increased their impact. But the numbers that you assess can be vague and, by themselves, hard to interpret. We have found useful measurements to fall into any of these three categories:

   • To sense-check/cover your bases: We have one type of outcome we assess# that we think is quasi-quantifiable in terms of $ provided to the movement. We take them with a huge grain of salt and apply large ranges for our estimates - but we can use these to assess whether we can cover or multiply our operating costs under optimistic assumptions/best guess assumptions/conservative assumptions. This way, the “harder to measure” forms of impact are “for free” and an “add-on” to the baseline impact, and subject to, e.g., a funder’s worldview.
   • For relative comparisons: Many numbers by themselves are hard to understand, but can tell you something in comparison to your past performance/other similar communities. At the time of writing, our Slack space has 832 weekly active users. By itself, this number doesn’t tell us much, but we do now know that we have ~200 more weekly active users compared to this time last year. It also puts us at a fairly high weekly active user count compared to other Slack communities in the EA ecosystem (the highest among the ones we could check!), and we have an unusually high weekly active user to total member ratio for a community of our size. Similarly, we know that we have fewer logged “High Impact Outcomes” than we did at this time last year, despite having a similar amount of FTEs, telling us that we may need to focus more on this (arguably more important) impact metric.
   • To test a specific hypothesis/your theory of change: While it is usually difficult to assess your overall impact and cost-effectiveness in community building, you can set out to “do things that reliably work and end up in impact” (i.e., a strong theory of change) and it will usually be easier to test for any specific point of your work whether or not it works out as intended, strengthening your overall theory of change and thus your path to impact. We have run a couple of experiments on our Slack space, around channel engagement strategies, onboarding, etc. - they usually didn’t produce game-changing results, but they at least confirmed that a specific action we took was working in the way we thought it would.
^{^}
A Note on Sustainability as Community Builder: Community building is emotionally demanding work that can easily lead to burnout if you’re not careful.
• Recognize burnout warning signs:
  • Dreading opening the community platform
  • Feeling resentful when members ask for help
  • Neglecting your own needs to serve the community
  • Feeling personally responsible for every member’s experience
  • Loss of boundaries between community time and personal time (this is a very common experience amongst community builders)
• Prevent burnout:
  • Set boundaries: Define work hours and stick to them. You don’t need to respond to every message immediately.
  • Share the load: Build that volunteer system earlier rather than later.
  • Take real breaks: Schedule time completely away from the community. Ensure you have someone to cover for you (this doesn’t mean they have to do everything you would typically do).
  • Remember your why: Regularly reconnect with your community members and the impact you’re having on them, not just the endless to-do list.
  • Recognize accomplishments (even the small ones): Don’t just check things off. Stop to celebrate or truly acknowledge that you’ve done something in line with your goals, even if it seems relatively small. At Hive, we have an internal #appreciations-wins channel where we encourage team members to share stories of impact and appreciation (and conveniently, it serves as a one-stop place for our impact tracking!)
  • Invest in peer support: Connect with other community builders who understand the unique challenges. You can set up regular calls or make sure to connect with them at relevant conferences and online.
• Plan for transitions: Document your processes and decision-making. You won’t lead this community forever, whether due to burnout, opportunity, life shifts, or strategic handoff. Make it possible for someone else to step in by:
  • Maintaining clear documentation of workflows, contacts and institutional knowledge.
  • Gradually sharing leadership with co-organizers or volunteers.
  • Creating systems that don’t rely solely on you being present.
• Consider Financial Sustainability: Beyond volunteer labor, how will your community sustain itself long-term?
  • Common funding models:
    • Grants: From EA funders, aligned organizations, or individual donors  (this is how Hive operates)
    • Fiscal sponsorship: Partner with an established organization to access its funding infrastructure (we used to have a fiscal sponsor)
    • Membership fees: Works for professional communities offering clear career value, common in for-profit industries (e.g., My Climate Journey Collective)
    • Sponsorships: Corporate or organizational sponsors who value access to your members (e.g., the not-with-us associated online community Nonprofit Hive runs with sponsorships)
    • Hybrid models: A Combination of the above
  • Most communities start with grants or volunteer labor, but thinking about long-term sustainability early helps you build and strategize appropriately and avoid a sudden shutdown when initial funding ends.
Show all footnotes