Community Research SOP

A community research SOP puts you inside the groups where your customers already talk to each other, and a useful round takes two to four weeks at about 30 minutes a day. A subreddit for nurses, a Discord server for indie game developers, a Facebook group for parents of twins or a monthly meetup of bakery owners holds years of honest questions and complaints. Read them properly, join in with real help, and the community gives you evidence, vocabulary and your first interviewees in one place.

What is the community research SOP and when should you use it?

Community research is deep, participatory research in a small number of groups where your target customers gather by choice. You read the history, learn the norms, contribute and then ask. It differs from social media listening, which tracks keywords across the open web without joining any conversation.

Use this SOP:

• When your customers are a defined group with their own forums, servers, associations or meetups.
• When cold outreach fails and you need trusted introductions.
• When you need the insider vocabulary that makes a landing page sound like it was written by one of them.

Communities also work as a recruiting base for customer interviews and, later, for customer co-creation sessions.

What do you need before you start?

• A customer profile from the persona development SOP, with job titles, hobbies or life situations that point to groups.
• A personal account with a real name and an honest bio. Fake personas get spotted and banned.
• A research log in Google Sheets (template below).
• Time every day: communities reward regular presence and punish one-off visits.

How do you run it, step by step?

1. Map 10 to 15 communities

Search Reddit, Discord server lists, Facebook and LinkedIn groups, Slack community directories, Meetup and trade association websites. Ask five people from your profile: "Where do you go when you have a question about [topic]?" For each community, log size, posts per week, rules and whether founders are welcome.

2. Pick the 3 to 5 most active

Choose by activity. A group of 2,000 with 30 posts a day beats a group of 80,000 where nothing happens. Mix at least two formats, such as a forum and a live meetup, so one platform's culture does not shape all your evidence.

3. Read a year of history before posting

Sort by "top posts of the year", read pinned posts, wikis and FAQ pages, and search for your problem keywords. Log every thread where someone describes the problem, a workaround or a request for a tool. Recurring questions that moderators keep answering are the strongest signal: they show a problem the community itself cannot solve.

4. Take part helpfully for two weeks

Answer questions where you know something, share resources and thank people. Do not mention your idea yet. Members check your history before they trust a research question, and a profile with only one post looks like a marketer.

5. Ask the moderators, then post one research question

Message a moderator, explain that you are researching a problem and ask whether a question post is welcome. Then post one open question about past behaviour: "How do you handle [situation] when [trigger] happens? What have you tried?" Say who you are and that you are not selling anything.

6. Turn replies into interviews

Reply to every answer, then message the 10 most detailed responders and ask for a 20-minute call. Offer to share a summary of what you learn with the whole group, and keep that promise.

7. Write the community brief

After 2 to 4 weeks, write one page per community: top problems with thread counts, exact phrases, workarounds mentioned, tools named and the people who agreed to talk.

Template you can copy

COMMUNITY MAP
Community | Platform | Members | Posts/week | Rules on founders |
Moderator name | Picked Y/N | Why

THREAD LOG
Date | Community | Link | Type (problem / workaround / tool request /
recurring question) | Exact words | Replies | Upvotes | Author contacted Y/N

RESEARCH QUESTION POST
Title: How do you handle [situation] when [trigger] happens?
"Hi all, I'm [name], [one line about you]. I'm researching how
[group] deal with [problem area] and I'm not selling anything.
When did this last happen to you, and what did you do?
I'll share a summary of the answers here in two weeks."

COMMUNITY BRIEF (per community)
Top 3 problems (thread count) | Best 5 quotes | Workarounds |
Tools named | People booked for interviews | Promise to follow up: date

Worked example: indie board game designers

This is an illustrative case with invented numbers. A founder thinks indie board game designers need help finding playtesters. She maps 13 communities and picks 4: a board game design subreddit, two Discord servers and a monthly designers' meetup in her city.

In a year of history she logs 64 relevant threads. Playtesting shows up in 12, but manufacturing quotes and minimum order sizes show up in 31, in all four communities, with the same question asked again and again: "Is this quote from a factory normal?". After two weeks of answering questions, she gets moderator permission and posts her question about the last time designers ordered a print run.

The post gets 47 replies, 26 with detail. Fourteen designers agree to a call. She changes the problem she is testing from playtesting to manufacturing quotes, keeps the same customer, and promises the subreddit a summary of typical quote ranges.

What should you measure, and when do you stop?

Stop when your thread log stops adding new problem types and you have booked enough interviews for your next round. If four weeks bring little activity, your customers gather somewhere else; go back to step 1 and ask more people where they ask questions.

Which mistakes should you avoid?

• Pitching on day one. It gets you banned and teaches communities to ignore you.
• Using a fake persona. Communities spot it, and the damage spreads to every group the members share.
• Picking groups by size. Activity and fit matter more than member counts.
• Taking without giving back. Share your summary, answer questions and credit the community when you publish.
• Ignoring offline groups. Trade meetups and associations hold customers who never post online.

How can AI speed this up?

Export threads you have permission to read, paste them into an AI assistant and ask it to list recurring questions, the problems behind them and the exact phrases used, with thread links. Ask it to draft your research question post in plain, community-friendly language. Never let AI write your replies to members; communities notice generic text, and trust is the whole asset. The F/MS audience and community module covers how to build your own community once you have found theirs.

Where does this SOP fit in your validation plan?

Community research runs early, next to interviews and listening.

1. Get the broad picture with social media listening.
2. Go deep in a few groups with this SOP.
3. Run customer interviews with the people you met.
4. Ask community moderators and long-time members for an expert interview, since they see hundreds of members' problems.

All 15 methods sit in the Customer Discovery category. The F/MS startup playbook blog has more on finding and working with early adopters.

FAQ

How do I find online communities for customer research?

Ask five target customers where they go with questions, then search Reddit, Discord, Facebook and LinkedIn groups, Slack directories, Meetup and trade association sites with their job titles and problem words.

Is it OK to do research in online communities?

Yes, if you follow the rules, say who you are, ask moderators before posting research questions and share what you learn. Never scrape private groups.

How long should I take part before asking a question?

About two weeks of genuine help. Members check your post history, and a profile with real contributions gets better answers.

What is the difference between community research and social media listening?

Listening tracks keywords across public platforms without taking part. Community research means joining a few groups, learning their norms and building relationships that lead to interviews.