Product Marketing Context
Establish what is known, interpreted, assumed, conflicting and missing before strategy or execution. Own intake, cross-domain synthesis, decision-specific readiness, research routing and integration of returned findings.
Boundaries and contracts
Read ../_shared/evidence-gaps.md, ../_shared/ssot-consumption.md and
../_shared/context-establishment.md before establishing context. Use its
inventory, provenance, contradiction, gap, handoff, return, readiness and proposal
contracts. Preserve source IDs, evidence labels, approval and limitations.
Do not perform specialist methodologies, select a market, generate personas, choose positioning/pricing, or author final messaging. Route those to their existing owners. Only ssot-context-loop writes canon, decisions, manifest and reconciler state. This is a bounded understanding workflow, not a GTM orchestrator.
Workflow
1. Resolve goal, scope and existing context
Identify the company/project, product, applicable region/segment and intended research question, decision or deliverable. Reuse supplied answers. Ask only for unresolved information that changes scope or the next consequential choice. A folder without a goal can be inventoried; do not declare readiness or prescribe an extensive study before the goal is clear.
Load applicable manifest-mapped SSOT and approved decisions first. Check
.agents/product-marketing-context.md and legacy .claude/product-marketing-context.md.
Keep existing content and full bootstrap profiles intact. On a scoped refresh,
change only relevant context and flag consequential dependencies. Do not silently
move a legacy file, replace approved claims, or infer approval from file existence.
2. Inventory mixed materials
Accept folders/files, pasted statements and accessible links: interviews/calls, surveys, CRM/win-loss, support/sales feedback, product docs/demos, decks, analyst reports, competitive research, pricing, positioning/messaging, personas, journey, SSOT and historical approvals. Codebase extraction and conversation remain input modes; neither is the default substitute for company evidence.
Create the source inventory before synthesis. Identify originals/derivatives, dates, source roles, scope, coverage, readability and reviewed portions. Inspect goal-relevant sources; keep unreadable/inaccessible/uninspected material visible. Do not fabricate extraction or claim comprehensive review from partial inspection. Reuse existing research and raw stores before gathering anything new.
3. Extract and synthesize candidate understanding
Record source-linked evidence separately from interpretations, assumptions, historical decisions, current approved decisions and unknowns. A deck establishes what it says, not that its claims are validated. User statements need scope and authority before they establish an approved decision. Preserve exact quotes.
Integrate existing domain findings; route raw material requiring domain analysis to its owner. Relate findings across domains without inventing their conclusions. For example, different buyer priorities by segment may be supported; selecting one segment remains market-entry-brief's decision. Classify disagreements using the shared contradiction register. Resolve apparent scope/time/definition conflicts only with demonstrated evidence; route unresolved factual/decision conflicts. Never favor newer or repeated wording over approved canon.
4. Diagnose decision-specific readiness
Identify what the actual goal needs. Assess support, authority and health separately for each requirement. Use exactly PROCEED, PROCEED PROVISIONALLY or HOLD THIS DECISION with the shared meanings. For provisional work, record the assumption, risk, allowed scope and validation trigger. A hold names the affected decision, gap, owner and closure criteria; continue unrelated work. Evidence sufficiency never grants human approval. Missing approval is a separate dependency. Avoid completeness scores or forcing irrelevant context sections.
5. Route consequential research gaps
Consolidate overlapping gaps into the smallest useful backlog, retaining context and lineage. Rank by what changes if wrong, uncertainty, urgency and effort. Retrieve → analyze existing evidence → clarify definitions → request internal facts → collect new research. Use the shared owner table and HND contract; one primary owner per question, contributors reuse the same evidence.
- Substantial research: synthesize first; route only unanswered domain analysis or consequential gaps, not a default new research program.
- Partial research: identify supported versus assumed/conflicting/missing context; route the gaps that change the user's goal and defer non-blocking ones.
- Starting from scratch: establish problem/product boundaries and constraints; route customer/category/competitive exploration before requiring an accepted market, complete persona or positioning. External market/category/opportunity research goes to market-intelligence; competitive landscape/facts/changes to competitor-profiling. Market-entry-brief owns entry evaluation and where/whether to enter. Then return evidence for market choice, personas within that scope, and strategy. Do not create circular prerequisites.
Handoffs do not authorize outreach or spending. If an owner is unavailable, keep its gap explicit and name the next action; do not silently absorb its methodology.
6. Integrate specialist returns
Check each return's scope, source locators, method, limitations and GAP completion criteria. Preserve supporting/opposing evidence and the original evidence/approval status. A report or increased citation count does not close a gap. Update answered, partial/inconclusive/unavailable work and reassess readiness. Follow up narrowly or state a stopping condition. Never loop until a preferred conclusion appears. Retain historical findings with supersession links instead of erasing them.
7. Assemble proposed PMM context
Read references/context-workspace-template.md when creating/updating the
workspace. Keep the familiar headings below; append registers and references to
existing domain artifacts. Reference covered approved context; retain full
uncovered bootstrap personas, committee/anti-persona detail and attribute sources.
Do not flatten them into summary rows. Missing applicable context stays Gap.
Prepare PRP records for supported portions with wording, destination, findings, assumptions/conflicts, domain owner, permitted use, pending validation and the human decision required. Do not wait for every non-blocking gap. Held choices remain unresolved, not disguised as settled proposals.
8. Review and hand off to SSOT
Show the inventory coverage, integrated context, readiness, consequential gaps, contradictions and concrete proposals for review. Save the provisional workspace when authorized, without changing canon. Corrections or saving do not approve it. No SSOT: hand proposals to ssot-context-loop /setup and await Gate 2. Existing SSOT: hand findings/proposals to /scan, Gate 1, /propose and Gate 2. Only SSOT persists approved changes. After an approved return, record replacement references while retaining pending/rejected/superseded proposal history. Tell the user which context is provisional and which approved references readers can use. Do not imply that every saved section is cleared for execution.
Sections to Capture
1. Product Overview
- One-line description
- What it does (2-3 sentences)
- Product category (what "shelf" you sit on: how customers search for you)
- Product type (SaaS, marketplace, e-commerce, service, etc.)
- Business model and pricing
2. Target Audience
- Primary audience plus alternates, each labeled Sourced, Assumed, or Gap. Keep alternates that lack proof; see
../_shared/segment-selection.md - Target company type (industry, size, stage)
- Target decision-makers (roles, departments)
- Primary use case (the main problem you solve)
- Jobs to be done (2-3 things customers "hire" you for)
- Specific use cases or scenarios
3. Personas
Reuse existing profiles from icp-and-buyer-personas; do not rebuild approved
personas during context setup or refresh. Keep the summary below for scanning.
Preserve full bootstrap profiles using that skill's attribute format, including
source/version, approval, provenance/confidence and gaps. With applicable SSOT,
reference its approved profiles instead of duplicating covered canonical facts.
Do not collapse profiles into a summary or message map when updating the context.
For B2B buying committees, capture:
- User, Champion, Decision Maker, Financial Buyer, Technical Influencer
- What each cares about, their challenge, and the value you promise them
4. Problems & Pain Points
- Core challenge customers face before finding you
- Why current solutions fall short
- What it costs them (time, money, opportunities)
- Emotional tension (stress, fear, doubt)
5. Competitive Landscape
- Direct competitors: Same solution, same problem (e.g., Calendly vs SavvyCal)
- Secondary competitors: Different solution, same problem (e.g., Calendly vs Superhuman scheduling)
- Indirect competitors: Conflicting approach (e.g., Calendly vs personal assistant)
- How each falls short for customers
6. Differentiation
- Key differentiators (capabilities alternatives lack)
- How you solve it differently
- Why that's better (benefits)
- Why customers choose you over alternatives
7. Objections & Anti-Personas
- Top 3 objections heard in sales and how to address them
- Who is NOT a good fit (anti-persona)
8. Switching Dynamics
The JTBD Four Forces:
- Push: What frustrations drive them away from current solution
- Pull: What attracts them to you
- Habit: What keeps them stuck with current approach
- Anxiety: What worries them about switching
9. Customer Language
- How customers describe the problem (verbatim)
- How they describe your solution (verbatim)
- Words/phrases to use
- Words/phrases to avoid
- Glossary of product-specific terms
10. Brand Voice
- Tone (professional, casual, playful, etc.)
- Communication style (direct, conversational, technical)
- Brand personality (3-5 adjectives)
11. Proof Points
- Key metrics or results to cite
- Notable customers/logos
- Testimonial snippets
- Main value themes and supporting evidence
12. Goals
- Primary business goal
- Key conversion action (what you want people to do)
- Current metrics (if known)
13. Positioning
Supplied by positioning-strategy as a referenced proposal or approved input. Approved mapped SSOT remains authoritative for covered positioning; this section supplies uncovered provisional context. Saving it does not approve strategy or modify canonical SSOT.
- Statement: one sentence in the form "[Product] is a [category + one short differentiating phrase] for [segment]."
- Derivation trace: Category, Differentiating phrase, Segment, and Trigger (the event that starts the search), each with its source and label
- Capability table: every differentiating capability, with what it does for the buyer, proof, buyer voice, and label
messaging-framework uses approved mapped SSOT capabilities where applicable and this table for uncovered provisional context. If the section is empty, approved SSOT inputs remain available; recommend positioning-strategy only for missing positioning.
14. Persona message map
Supplied by icp-and-buyer-personas as a referenced proposal or approved input. messaging-framework uses approved mapped SSOT persona messaging where applicable, and this map for uncovered provisional context. Preserve evidence labels and proposal/approval status; the map does not change canon.
- Persona: role-based name with the buying role in brackets, such as (champion) or (financial buyer)
- What matters to them: 1 to 2 short phrases
- Message: the core message in 1 to 2 sentences
- Proof label: Sourced, Assumed, or Gap for the proof behind the message
Output rules
Open internal output with Inputs and assumptions. End with ranked Evidence gaps, including all assumptions/gaps and suggested owners/milestones when unknown. No invented proof, source versions, quotes or approvals. No em dashes.
