---
name: "generate-business-model"
description: "Generate a justified, research-backed business model for an existing web or mobile project. Use when the user asks to define, challenge, compare, or create a business model, monetization model, pricing strategy, Business Model Canvas, go-to-market economics, or investor/client-ready business model report for a digital product, SaaS, marketplace, mobile app, AI app, ecommerce, subscription app, community, platform, or service. The skill guides project analysis, targeted user questions, web research, evidence scoring, country/sector differences, model selection, justification of tradeoffs, and creation of a PDF report authored by the AI agent."
---

# Generate Business Model

## Outcome

Produce a defensible business model for an existing project, usually web or mobile, with:

- project diagnosis and assumptions
- focused questions to the user when data is missing
- sourced research and benchmark analysis
- 2-4 candidate models compared
- final recommended model with justified choices
- revenue logic, pricing, KPIs, risks, experiments, and next steps
- a PDF report generated by `pdf-generator` from an AI-authored structured payload

## Core Workflow

1. **Frame the project**
   - Identify product type, users, buyer, geography, sector, current traction, existing monetization, acquisition channels, and constraints.
   - If a repository, product brief, Notion page, app store listing, website, analytics export, or pitch deck is available, inspect it before asking questions.
   - If context is insufficient, ask only the next most useful questions from `references/question-bank.md`, grouped by priority.

2. **Build the hypothesis map**
   - Use `references/methodology.md`.
   - Translate the project into hypotheses across desirability, feasibility, viability, and defensibility.
   - Mark unknowns explicitly; never fill key commercial facts with invented numbers.

3. **Research the market**
   - Browse when facts may be current, market-specific, country-specific, or benchmark-dependent.
   - Start with primary and high-authority sources: company filings, official platform docs, regulator docs, payment docs, benchmark reports, academic articles.
   - Use `references/source-standards.md` to score and cite sources.

4. **Select model candidates**
   - Use `references/model-patterns.md`.
   - Always compare at least 2 plausible models unless the user explicitly requests one.
   - Include country/sector/store constraints where relevant: payment methods, VAT/tax, app store billing, local purchasing power, regulation, trust, and platform dependency.

5. **Recommend and justify**
   - Justify every major choice: customer segment, value proposition, revenue stream, pricing metric, channel, retention mechanism, cost structure, key partners, and KPIs.
   - Label claims:
     - `Verified fact` for directly sourced facts.
     - `Derived recommendation` for your synthesis from evidence.
     - `Assumption` for unverified but necessary assumptions.
   - Separate "best current model" from "experiments required before committing."

6. **Generate the PDF via `pdf-generator`**
   - Write the final report content using `references/report-template.md`.
   - Build the structured payload described in `references/report-template.md`.
   - Invoke `$pdf-generator` and pass the payload. The PDF must be generated and delivered by that skill, not by this skill directly.
   - After `$pdf-generator` returns, relay `PDF_PATH`, `HTML_PATH`, `PAGES`, and `STATUS` to the user from inside the `generate-business-model` response.
   - If `$pdf-generator` is unavailable, do not silently substitute another renderer. Tell the user the PDF generator skill is missing and offer to save Markdown or use `scripts/markdown_to_pdf.py` as an explicit fallback only after user approval.

## Research Rules

- Prefer fewer strong sources over many weak summaries.
- Use at least 5 sources for full reports:
  - 1 methodology source (Strategyzer, Bpifrance, HBR/Lean Startup, academic)
  - 1 sector/model benchmark source
  - 1 platform/payment/regulatory source when web/mobile distribution matters
  - 1 or more comparable company examples when available
- For business models copied from successful companies, use primary sources when possible: SEC filings, annual reports, official pricing pages, official developer docs.
- Do not present benchmarks as universal truth. Segment by sector, country, platform, and stage.

## Questioning Rules

- Ask questions only when the answer cannot be inferred from project artifacts or research.
- Ask a maximum of 7 questions at once.
- Prioritize questions that change the model choice:
  - Who pays?
  - What value is repeated?
  - What usage creates variable cost?
  - What market/country is first?
  - What traction exists?
  - What current acquisition channel works?
  - What legal/platform constraint applies?

## Report Quality Bar

A good answer must include:

- a short executive summary
- candidate model comparison table
- final Business Model Canvas
- pricing and monetization logic
- country/sector adaptations
- source-backed justifications
- metrics to track
- validation experiments
- clear unresolved assumptions
- a generated PDF path/link/status returned by `pdf-generator`

Avoid:

- generic "subscription is best" answers
- unsupported revenue forecasts
- ignoring app store/payment/tax constraints
- confusing business plan with business model
- optimizing only for investors instead of business viability

## Resource Map

- `references/methodology.md`: step-by-step analytical method.
- `references/question-bank.md`: prioritized questions for the user.
- `references/model-patterns.md`: model selection matrix for web/mobile projects.
- `references/source-standards.md`: research, scoring, and source examples.
- `references/report-template.md`: final report structure and `pdf-generator` payload contract.
- `scripts/markdown_to_pdf.py`: emergency fallback only when `pdf-generator` is unavailable and the user approves a local PDF.