A mastermind group is five or six people with no shared boss, no shared payroll, and no reason to keep showing up except the value of the meeting itself. That is the whole fragility of the format: the moment sessions stop being clearly worth two hours, people stop protecting the slot. Nobody announces this. Attendance just softens.
Groups that last for years are not the ones with the best members. They are the ones with the most structure — which is the opposite of what most founders expect when they start one.
Why do most mastermind groups dissolve by month eight?
Drift, in a predictable sequence. First the format loosens: the hot seat becomes a group chat, the group chat becomes a catch-up. Then the commitments stop being checked, so they stop being made seriously. Then one busy member misses two sessions and discovers nothing was lost by missing them — and once that is discovered, it cannot be un-discovered.
Notice what is absent from that sequence: conflict, bad members, bad intentions. Groups almost never die of anything dramatic. They die of sessions that stopped having a spine.
What format actually holds up?
The boring one, executed every single time. For a two-hour monthly session of five or six people:
- Check-in round: each person, three minutes, headline of their month against what they said last time. Timed, no discussion.
- Commitment review: read last month's commitments out loud, each marked done or not done, one sentence of why. Ten minutes.
- Hot seats: two members, thirty-five minutes each — ten to present the problem, twenty of questions before any advice, five of “what I'm taking from this.”
- Commitments: everyone states one, with a date. Written down, read back. Ten minutes.
Two hot seats a session means everyone gets deep help roughly every quarter, which is enough — because the check-ins and commitment reviews are where the compounding actually happens.
Why do questions-before-advice matter so much?
Because the natural failure mode of smart, generous people is to start advising ninety seconds into hearing a problem — and the advice will be excellent solutions to the wrong problem. The presenting problem (“I need to hire a salesperson”) is rarely the real one (“I hate selling and want permission to stop”).
Twenty minutes of questions with a hard ban on suggestions changes what surfaces. Not “have you tried…” disguised as a question — real ones: What have you already tried? What would make this problem disappear without solving it? What are you optimising for this year? The advice that follows twenty minutes of that is worth ten times the advice that follows ninety seconds.
What should the group write down?
Three things, kept in one shared document that gets read at the start of every session rather than filed and forgotten: what each person committed to, what each hot seat concluded, and what themes keep coming back.
The last one is the quiet superpower of a long-running group. When someone brings the same flavour of problem for the third time — a different hire, the same avoidance; a different launch, the same perfectionism — a group with a record can see it and say it. That observation is worth more than any single session's advice, and it is only available to groups that keep score. Without the record, every problem arrives new, and the group stays permanently shallow.
How do you handle the people problems?
Set the norms before there is a problem, because retrofitting rules onto a specific person never works. The three that matter:
- Attendance is binary: you attend or you leave the group. Two unexcused misses is a conversation; three is an exit. Harsh-sounding, but it is what keeps the slot sacred for everyone else.
- Confidentiality is absolute and stated out loud once a quarter. The first time something said in the room travels, the room is dead — people will keep attending and stop saying anything real.
- Facilitation rotates, or nobody facilitates honestly. A permanent facilitator becomes the owner, the owner develops stakes, and members stop telling the owner things. Rotating the role monthly keeps the group's neutrality — the timekeeper this month was on the hot seat last month.
The rotation has one real cost: quality varies. Some members run a tight session; some let the check-in round sprawl to forty minutes. Groups survive variable facilitation. They do not survive absent facilitation.
Common questions
How many people should a mastermind group have?
Five or six. Four is fragile — one absence and it's a coffee chat. Seven or more means hot seats come round too rarely to hold anyone's commitment, and check-ins eat the session.
How often should a mastermind group meet?
Monthly for two hours is the durable default. Fortnightly works for groups in the same trench (e.g. all pre-launch founders) but burns out generalist groups. Quarterly is too rare to build the trust the hot seat needs.
Should a mastermind group be paid or free?
Peer groups can be free if the attendance rule has real teeth. If nobody is willing to enforce it, a modest stake everyone pays into — forfeited on no-shows, donated at year end — does the enforcing impersonally.
What kills a mastermind group fastest?
Unchecked commitments. The moment saying “I'll do X by the 15th” carries no follow-up, the sessions become social, and busy people cut social before they cut useful. The commitment review is the spine — protect it above everything else.
I'm building Facilitable, a facilitator for groups that meet regularly — it runs the rounds and hot-seat timers, keeps every commitment on record, and notices the themes that keep coming back, so the rotation never depends on who's good at facilitating. It isn't ready yet. If you'd like to see it when it is: