Meetings without minutes are parties. Meetings without state changes are mispriced attention.
The Job
When a team needs to align on decisions, surface friction, or coordinate action across ventures, help them run a structured meeting that produces at least one traceable artifact — a problem identified, a standard proposed, a PRD updated, or a scoreboard metric moved.
| Trigger Event | Current Failure | Desired Progress |
|---|
| Team sync needed | Conversation drifts, no decisions captured | Every meeting produces a routed state change |
| New friction surfaced | Ad-hoc discussion, forgotten by next week | Friction logged as a Problem with owner |
| Decision required | Authority unclear, commitment decays post-meeting | Decision recorded with bet, owner, and due date |
| Creative collision | Serendipity wasted without follow-through | Collision captured and matched to venture opportunity |
Collision Algorithm
The highest-value meetings are creative collisions — unplanned collaborations between people in complementary zones of proximal development. The system should match people who can be each other's MKO (More Knowledgeable Other).
| Without System | With System |
|---|
| Collisions are accidental | Collisions are orchestrated by capability gap |
| No follow-through on sparks | Every collision logs a friction or opportunity |
| Same people talk to same people | ZPD matching surfaces complementary pairs |
| Value of meeting is unmeasured | Collision-to-commitment conversion tracked |
Tight Five Agenda
Every meeting invite must answer these five questions on one screen. If it cannot, it is a party — allowed for culture, forbidden for decisions.
| # | Question | Maps To |
|---|
| 1 | Why does this matter? | Purpose / venture |
| 2 | What truths guide us? | Standards / Principles |
| 3 | What do we control? | Platform / current constraints |
| 4 | What do we see others don't? | Questions / perspectives surfaced |
| 5 | How do we know it's working? | Scoreboard metric to move |
Feature / Function / Outcome
| # | Feature | Function | Outcome |
|---|
| 1 | Tight Five agenda builder | Generate meeting invite from 5 questions with linked context | Every meeting starts with explicit purpose |
| 2 | Meeting type classifier | Tag as Collision / Discovery / Debate / Accountability / Decision | Right protocol applied to right meeting |
| 3 | Structured minutes capture | Record decisions, bets, owners, due dates during meeting | Commitments are traceable, not forgotten |
| 4 | Output router | Route each minute item to Problems, Standards, PRDs, or Scoreboard | Meeting outputs land in the right system artifact |
| 5 | Collision matcher | Match people by ZPD gap and complementary capability | Higher quality creative collisions |
| 6 | Decision journal integration | Log decisions with confidence level and falsification criteria | Decision quality compounds over time |
| 7 | Cadence templates | Daily/weekly/monthly/quarterly meeting presets | Consistent rhythm without reinventing agendas |
| 8 | Debrief protocol | Post-meeting JTBD mapping and VVFL stage tagging | Every meeting feeds the feedback loop |
| 9 | Attendance gating | Require authority-level declaration for decision meetings | Only decision-makers in decision meetings |
| 10 | Follow-through tracker | Surface uncommitted actions and overdue commitments | Commitment doesn't decay |
Meeting Cadence
| Cadence | Meeting Type | Input | Output | Feeds |
|---|
| Daily | Decision | Question + Problem + Action | One concrete move per venture | PRD commissioning |
| Weekly | Accountability / Decision | Feature + Function + Outcome review | Updated commissioning state | Scoreboard |
| Monthly | Discovery / Debate / Decision | Pilot + Proof + Pricing iteration | Evidence-linked business dev updates | Business development |
| Quarterly | Standards | Protocol + Standard + Platform review | Retire, promote, or refactor protocols | Standards |
Minutes Schema
Every meeting minute is a micro-commissioning artifact:
| Field | Purpose | Links To |
|---|
| Meeting type | Collision / Discovery / Debate / Accountability / Decision | Cadence template |
| Venture | Which mycelium venture this serves | Mycelium |
| Tight Five answers | The five questions answered before meeting | Agenda builder |
| Attendees + authority | Who was there and what they could decide | Attendance gate |
| Decisions | What was decided, with confidence and owner | Decision journal |
| Bets | Falsifiable predictions made | Predictions |
| Routed outputs | Where each action item lands in the system | Output router |
| VVFL stage | Which loop stage this meeting advanced | VVFL |
Success Criteria
| Type | Criterion | Threshold |
|---|
| Functional | Create meeting with Tight Five agenda | Under 5 minutes |
| Functional | Capture and route minutes to system artifacts | 100% of decision items routed |
| Functional | Match two people by ZPD gap for collision | Under 10 seconds |
| Outcome | Meetings producing at least one state change | >= 90% |
| Outcome | Decision follow-through within agreed timeframe | >= 80% |
| Outcome | Reduction in repeated alignment meetings on same topic | >= 50% reduction |
Commissioning
| Component | Schema | API | UI | Tests | Status |
|---|
| Tight Five agenda builder | Pending | Pending | Pending | Pending | 0% |
| Meeting type classifier | Pending | Pending | Pending | Pending | 0% |
| Minutes capture | Pending | Pending | Pending | Pending | 0% |
| Output router | Pending | Pending | Pending | Pending | 0% |
| Collision matcher | Pending | Pending | Pending | Pending | 0% |
| Decision journal link | Pending | Pending | Pending | Pending | 0% |
| Cadence templates | Pending | Pending | Pending | Pending | 0% |
| Follow-through tracker | Pending | Pending | Pending | Pending | 0% |
Risks + Kill Signal
| Risk | Mitigation |
|---|
| Over-structuring kills spontaneity | Keep collision meetings deliberately loose; structure only the capture |
| Adoption friction from agenda requirements | Start with decision meetings only; expand when the value is obvious |
| Minutes become bureaucratic overhead | AI-assisted capture; one-click routing; never more than 5 minutes post-meeting |
| Collision matching feels forced | Opt-in matching; surface suggestions, don't mandate |
Kill signal: If teams consistently skip the Tight Five agenda AND follow-through rates don't improve within 60 days, the protocol is adding friction without value.
Context
Questions
What is the single most important thing that must be true for meetings to succeed — and how would you know if it weren't?
- Which user struggle does this PRD solve most directly — and is that the same struggle that would drive adoption?
- What is the minimum viable version of this feature that would still produce the core outcome?
- If this PRD were never built, what would users do instead — and is that workaround good enough?
