| name | stg-designing-solutions |
| description | Designs solution architecture, selects growth model, assembles positioning, maps features to problems, and defines MVP scope. Use when constructing Solution Design section during BUILD phase 2. |
| serves | strategist |
| domain | solution-design |
| affects | solution-design-section |
| depends-on | none |
| produces | Solution Design section in register format |
Solution Design
Design solution architecture: select growth model, assemble positioning, map features to problems, define MVP scope, design growth loops. Output is the Solution Design section of the register -- a derived design artifact constrained by all three hypotheses, not a hypothesis itself.
Procedure
Step 1: Load Hypothesis Context [S]
Read: problem, segment, unit economics hypotheses. Also read competitive analysis (gaps, alternatives) for positioning assembly.
Extract constraints each hypothesis places on the solution:
- Problem -> what must be solved (top problems by severity)
- Segment -> buyer type (user=buyer or committee), ACV range, time-to-value expectations
- Unit Economics -> cost constraints (gross margin target, COGS limits)
Extract positioning inputs:
- Problem claim -> the problem statement for the positioning template
- Segment claim -> the target for the positioning template
- Competitive analysis -> alternatives and gaps for the "unlike" clause
- Aha moment (assembled in Step 5) -> unique capability clause
Produce: hypothesis constraints table + positioning inputs.
Gate: constraints_extracted: bool -- at least one constraint from each hypothesis documented, positioning inputs identified.
- Pass: Step 2.
- Fail: If a hypothesis is missing, note the gap. Proceed with available constraints, flagging reduced confidence.
Step 2: Select Growth Architecture [R]
Score buyer characteristics against decision matrix:
| Factor | PLG | Network | Traditional | Source |
|---|
| ACV | < $5K | < $5K | > $25K | Segment hypothesis |
| Buyer type | User = buyer | User = buyer | Committee | Segment hypothesis |
| Time-to-value | < 1 day | < 1 day | Weeks+ | Product nature |
| Collaboration | Optional | Inherent | Either | Product nature |
| Shareability | Medium | High | Low | Product nature |
For ACV between $5K-$25K: Hybrid. Document which factors pulled in which direction.
Selection is T1/T2 (derivable from segment characteristics).
Produce: growth architecture selection with rationale [T1/T2].
Gate: architecture_selected: bool -- selection justified by segment characteristics, not asserted.
- Pass: Step 3.
- Fail: If factors conflict (e.g., low ACV but committee buyer), flag the tension and select with documented reasoning.
Step 3: Assemble Positioning [K-grounded]
Grounded in: Problem claim, Segment claim, competitive analysis gaps, growth architecture selection.
Assemble the positioning statement from hypothesis outputs -- this is composition, not hypothesis construction. Every clause traces to an existing hypothesis or research output:
- Target: from Segment hypothesis claim
- Problem: from Problem hypothesis claim
- Category: determine the category the product occupies in the buyer's mind. If category is ambiguous or contested, mark as [GOVERNOR_DECISION] for escalation.
- Differentiator: from competitive analysis gaps -- what the product does that alternatives do not
- Unlike [alternative]: from competitive analysis -- the primary alternative the target currently uses
- Unique capability: from aha moment (forward reference -- finalize after Step 5 defines aha moment, revise positioning if needed)
Produce: assembled positioning statement + category framing decision.
Note: the "unique capability" clause depends on the aha moment defined in Step 5. Write a draft here using competitive gaps. Revise after Step 5 if the aha moment provides a stronger unique capability clause.
Gate: positioning_assembled: bool -- positioning statement has all six clauses filled (unique capability may be draft), category framing documented.
- Pass: Step 4.
- Fail: Identify which clause cannot be filled. If target or problem clause is empty, the upstream hypothesis is missing -- halt and report.
Step 4: Map Features to Problems [K-grounded]
Grounded in: problem hypothesis (top problems), segment characteristics (jobs addressed).
For each top problem, identify features that solve it. For each feature, classify:
| Job Dimension | What It Means | Example |
|---|
| Functional | What they need accomplished | "Generate reports in 2 minutes" |
| Emotional | How they want to feel | "Feel confident in data accuracy" |
| Social | How they want to be perceived | "Be seen as data-driven leader" |
Flag if all features are functional-only: "Technical founders design for the functional job and ignore emotional and social jobs."
Prioritize each feature:
- MVP -- minimum for aha moment
- POST_MVP -- important but not for first value
- FUTURE -- nice to have or expansion
Produce: feature map with job dimensions and priorities [T2].
Gate: features_mapped: bool -- every feature traces to a stated problem, at least one non-functional job dimension addressed, priorities assigned.
- Pass: Step 4.
- Fail: Features that do not trace to any stated problem must be removed or the problem must be added to the hypothesis. Every feature must trace.
Step 5: Define MVP Scope [K-grounded]
Grounded in: feature map, growth architecture.
MVP = minimum set that delivers the aha moment.
Aha moment definition: the specific experience where the user says "this solves my problem." Must be concrete and experiential:
- Good: "Generate a complete design system from 3 inputs in 90 seconds"
- Bad: "See value in the product", "experience efficiency"
Time-to-value target:
- PLG: must be < 1 day from signup
- Network: must be < first interaction with other user
- Traditional: must be < first sales call follow-up
Excluded features: list each with why excluded and when to add.
Mode overlay:
- BOOTSTRAP -> smallest possible MVP for fastest revenue
- VENTURE -> MVP can be slightly larger if it enables growth loop
Produce: MVP scope with aha moment and time-to-value [T2 for scope, T3 for timing effectiveness].
Gate: mvp_defined: bool -- aha moment is specific and experiential, time-to-value target stated, excluded features have rationale.
- Pass: Step 6. After defining aha moment, revisit Step 3 positioning -- revise "unique capability" clause if aha moment provides a stronger formulation.
- Fail: If aha moment is vague, rewrite with specific experience, observable outcome, and timeframe. If MVP has >5 features, challenge: which single feature delivers the aha moment? Start there.
Step 6: Design Growth Loops [K-grounded]
Grounded in: growth architecture, feature map, segment characteristics.
For selected architecture, define growth mechanisms:
| Architecture | Loop Pattern | Requires |
|---|
| PLG | Product usage -> sharing -> new signups -> more usage | Shareable output, invite mechanism |
| Network | User joins -> invites others -> network value increases | Collaboration features, density threshold |
| Traditional | Sales close -> implementation success -> referral | Customer success process, case studies |
Each loop: mechanism, what it requires (features, scale), tier label.
Produce: growth loop definitions [T2].
Gate: loops_designed: bool -- at least 1 growth loop with mechanism, requirements, and tier label.
- Pass: Step 7.
- Fail: If no natural growth loop exists for the architecture, document: "Growth loop is weak -- customer acquisition will depend primarily on paid channels." This is a finding.
Step 7: Write Solution Design Section [S]
Write complete Solution Design section in register format:
- Growth Architecture: {PLG | Network | Traditional | Hybrid} with rationale
- Architecture Rationale: ACV, buyer type, time-to-value, collaboration, selection reasoning
- Positioning: assembled statement with all six clauses, category framing, category rationale
- Feature Map: table with feature, solves problem, job dimension, priority, tier
- MVP Scope: included features, aha moment, time-to-value target, excluded features with rationale
- Growth Loops: loop name, mechanism, requirements, tier
- Constraints from Hypotheses: table with constraint from each hypothesis (Problem, Segment, Unit Economics) and what to re-evaluate if hypothesis changes
- Adequacy Criteria: specific, measurable criteria for solution adequacy (not kill conditions -- this is a design artifact, not a hypothesis)
- Last Updated: date
Produce: complete Solution Design section.
Gate: section_written: bool -- all fields populated, positioning statement complete, constraints reference all three hypotheses.
- Pass: Done.
- Fail: Fix missing fields. Ensure constraints table references all three hypotheses. Ensure positioning statement has all six clauses finalized (unique capability revised after aha moment).
Quality Criteria
- Growth architecture selection justified by segment characteristics (not asserted)
- Positioning statement assembled from hypothesis outputs (not invented independently)
- Every positioning clause traces to a hypothesis claim or competitive analysis output
- Category framing decision documented (with governor escalation if ambiguous)
- Every MVP feature traces to a problem in the Problem hypothesis
- At least one feature addresses emotional or social job (not all functional)
- Aha moment is specific and experiential (not a vague benefit)
- Constraints from all three hypotheses documented
- Excluded features have rationale and "when to add" guidance
- Growth loops are mechanistically described (not just named)
Failure Modes
| Mode | Signal | Recovery |
|---|
| Feature completeness before launch | MVP contains >5 features or includes "platform" / "ecosystem" language | Flag technical founder pattern. Ask: which single feature delivers the aha moment? Start there. Everything else is POST_MVP until customer evidence demands it |
| Aha moment is vague | "Users see value", "teams get results", "experience efficiency" | Rewrite as specific experience with observable outcome and timeframe. "Complete X task in Y minutes that previously took Z hours" |
| Growth architecture mismatch | PLG selected but ACV >$25K, or Traditional selected but ACV <$2K | Architecture-ACV misalignment. Either the pricing is wrong or the architecture is wrong. Flag for resolution in economics hypothesis |
| Positioning invented rather than assembled | Positioning statement contains claims not traceable to any hypothesis or competitive analysis | Every clause must trace to a source. Target = Segment claim. Problem = Problem claim. Alternative = competitive analysis. If a clause has no source, it is fabrication |
| Solution not constrained by hypotheses | Features that do not trace to any stated problem; growth loops that assume segment behaviors not in evidence | Every feature must trace to the Problem hypothesis. Every growth loop must be consistent with Segment characteristics. Remove untraced elements or add the tracing |
Boundaries
In scope: Growth architecture selection, positioning assembly from hypothesis outputs, feature-to-problem mapping with JTBD dimensions, MVP scope definition with aha moment, growth loop design, hypothesis constraint documentation, adequacy criteria.
Out of scope: Pricing design (stg-designing-pricing), economics calculation (stg-calculating-economics), problem scoring (stg-scoring-problems), competitive analysis (stg-analyzing-competition).