MANDATORY PREPARATION

Invoke /product-thinking — it contains the Context Gathering Protocol and the AI Slop Test. Follow the protocol before proceeding.

Mindset

A workshop is not brainstorming. Brainstorming produces 40 ideas and zero decisions. A workshop starts from grounded inputs — what users actually do, what competitors actually ship, where value actually flows — and converges on opportunities worth pursuing.

This is a conversation, not a monologue. Push back. Ask "why would someone care?" Challenge assumptions. The best workshops produce two or three sharp ideas, not twenty vague ones.

Context Pull

Load all available intelligence:

Product context. Read .acumen.md — strategy, positioning, stage, constraints.
Personas. Read .acumen/personas.md — who we serve, their behaviors, workarounds, pain points.
Features. Read .acumen/features.md — what exists today, gaps, decay.
Value chain. Read .acumen/value-chain.md — the end-to-end workflow, where the product sits, extension points per persona. This is the foundation for Lens A.
Competitors. Read .acumen/competitors.md — what the market offers, feature parity traps, moats.
Data sources. Check .acumen/sources.md — pull real usage data, feedback, support patterns if available.
Prior reports. Check .acumen/reports/ for recent /diagnose or /measure outputs that triggered this workshop.

Workshop Modes

The workshop has three input lenses. Use all three by default, or focus on one if the user specifies a direction.

Lens A: Value Chain Analysis

Start from .acumen/value-chain.md — the persisted workflow map. Do not rebuild from scratch.

Review the value chain. Load the existing map. Verify it still reflects reality for the workshop topic. If steps are missing or stale, update the map as part of the workshop.
Focus on extension points. The value chain already identifies where personas switch tools, do manual work, or lose time. Zoom in on the ones relevant to this workshop's topic.
Evaluate each extension:
- Does it reinforce the core thesis or dilute it?
- Would users expect your product to do this?
- What's the effort-to-value ratio?
- Does it deepen a moat (data, switching costs, workflow ownership)?

Lens B: Competitive Feature Scan

Pull from .acumen/competitors.md:

What do competitors offer that we don't? List features, not marketing claims.
For each: is it a real gap or a parity trap?
- Does our primary persona actually need this?
- Would building it serve our thesis or chase theirs?
- What would we stop doing to build this?
What do we offer that competitors don't? These are differentiation opportunities to double down on.
What is nobody building? White space in the market. Jobs that are underserved across all players.

Lens C: User Signal Synthesis

Pull from configured feedback sources (support, NPS, interviews, analytics):

What are users asking for? Group requests by theme, not by individual ticket.
What are users complaining about? Distinguish usability complaints (the feature is hard to use) from value complaints (the feature doesn't solve my problem).
What are users working around? Workarounds reveal unmet needs. A user who exports data to Excel to do analysis is telling you something.
What are users NOT doing? Features with low adoption. Flows with high drop-off. Silence is signal.

If no feedback sources are configured, say so explicitly and work from the other two lenses.

Persona Mode

When exploring opportunities, think AS specific personas from .acumen/personas.md. For each relevant persona:

Walk through the opportunity from their perspective
What would excite them? What would they ignore?
Would they switch from their current workaround?
How does this fit into their daily workflow — or does it create a new one they need to adopt?

This replaces generic "users would..." with grounded persona-specific reactions.

Workshop Flow

Frame the space. State the opportunity area and why we're exploring it. Reference the trigger — a /diagnose finding, a /measure insight, a competitive move, or a user request pattern.
Run the lenses. Apply all three (or the user's chosen focus). Surface findings, not just data.
Generate opportunities. Each opportunity must have:
- The bet: what we believe will be true
- Who it serves: specific persona(s)
- Evidence: which lens surfaced it and what supports it
- Thesis fit: does it reinforce or extend the product thesis?
- Moat contribution: does it build defensibility?
Converge. Rank opportunities by a simple 2x2: evidence strength (how confident are we this matters?) vs. thesis fit (how well does this align with where we're going?). Top-right quadrant is where to focus.
Suggest next steps. For the top 2-3 opportunities, recommend:
- /increment to scope it
- /roadmap to sequence it with other work
- /measure to define how we'd know it worked

Output Format

Workshop: [Topic]

Trigger: What prompted this workshop (diagnosis, metric check, competitive move, intuition)

Value Chain Map

Visual or table showing the persona's workflow and where the product sits today.

Opportunity Board

#	Opportunity	Persona	Evidence	Thesis Fit	Moat Impact	Source Lens
1			Strong/Medium/Weak	Direct/Extends/Neutral	Builds/Neutral/None	A/B/C

Top Opportunities (detail)

For each top 2-3:

[Opportunity Name]

The bet: [one sentence]
Who it serves: [persona name and why]
Evidence: [specific data, feedback, or competitive signal]
Current workaround: [what users do today]
Thesis fit: [how it connects to product strategy]
Risks: [what could go wrong or why this might be wrong]
Next step: /increment, /roadmap, or /measure

Parked Ideas

Ideas that surfaced but don't meet the bar right now. Keep them so they don't get lost.

What We Deliberately Ignored

Opportunities we considered and rejected, with reasoning. This prevents the same ideas from resurfacing without new evidence.

Save workshop output to .acumen/reports/workshop-[topic]-[date].md.

If the workshop revealed new workflow steps, extension points, or persona paths not in .acumen/value-chain.md, update the value chain map.

workshop