business-plan

Produces a full business plan: a comprehensive, standalone planning document that states what the business is, who it serves, how it competes, how it operates, and how it makes and spends money — written so a lender, investor, or partner can decide whether to fund or work with it from this document alone. Its center of gravity is the Financial Plan & Projections section's disclosed assumptions — a plan that presents projected revenue or profitability without disclosing the assumptions behind them (pricing, unit volume, cost basis, growth rate) is not a conformant business plan, no matter how polished the narrative reads.

This genre follows the widely-used SBA/SCORE-style business-plan structure — conventional practice among lenders and business counselors, not a single codified format standard — supplemented by Lean Canvas's problem-solution framing for early-stage ventures. State this plainly rather than claiming "SBA-approved" or "SBA-required" format; the SBA publishes guidance and a checklist, not a certifying format standard.

Pattern (industry: SBA/SCORE-style business plan, Lean Canvas problem framing)

1. Executive Summary — a one-to-two page snapshot: what the business does, who it serves, how it makes money, its traction to date, and (if seeking financing) how much and why. Written last, placed first.
2. Company Description — legal entity, structure, formation date and state, ownership, mission, and the problem the company exists to solve (Lean Canvas's Problem block: the target customer's top 1-3 pain points) and its Solution.
3. Products & Services — what is offered, pricing, and the Unique Value Proposition: the single, clear message that states why the business is different and worth buying from.
4. Market Analysis — target market size and segments, customer profile, and buying criteria, supported by cited research, not assertion.
5. Competitive Analysis — named competitors, their strengths and weaknesses, and this business's Unfair Advantage (something a competitor cannot easily copy or buy).
6. Marketing & Sales Strategy — positioning, channels to reach customers, pricing/distribution strategy, and the sales process.
7. Operations Plan — how the business runs day to day: location, suppliers, facilities/equipment, and key operational milestones.
8. Management & Organization — legal structure, the management team and their relevant qualifications, org chart, and staffing plan.
9. Financial Plan & Projections — see Rules, below; this section is the genre's mandatory core.
10. Funding Request (only when financing is sought) — amount requested, use of funds, and terms sought. Omitted cleanly (not stubbed with placeholder numbers) for a self-funded or internal-strategy plan.
11. Appendix — supporting detail: resumes, permits/licenses, leases, letters of intent, detailed financial schedules.

Rules that keep it a business plan

• The Financial Plan & Projections section's disclosed assumptions are mandatory, not optional matter. At minimum: a 3-statement projection (income statement, cash flow statement, balance sheet) covering a 3-year horizon — monthly detail for year 1, quarterly for year 2, annual for year 3 is the investor-ready convention — plus a break-even analysis, and every material assumption stated explicitly (unit pricing, volume/growth rate, cost basis, hiring plan). An unlabeled number with no stated assumption behind it is a defect, not a simplification.
• Every external market, competitive, or industry claim traces to a cited source or an explicitly labeled assumption — never a bare, unsourced statistic. Where the drafting context and any available research (e.g. a prior market-research-report or mif-corpus search hit) leave a gap, gather it live (web research) or elicit it from the author; if neither resolves it, state the gap and the assumption used to bridge it in the same sentence as the figure it supports. This is how the genre earns reader confidence without asserting false precision.
• The Funding Request section is genre-conditional: include it, fully specified (amount, use of funds, terms), only when the plan seeks financing. Do not stub it with invented numbers for an internal or self-funded plan — omit the section and say why in the Executive Summary instead.
• Render every financial or market chart within Mermaid's range as a fenced mermaid code block — xychart-beta for revenue/cash-flow trajectories, pie for market segmentation or use-of-funds breakdown of five or fewer segments, quadrantChart for competitive positioning if used — never ASCII art, or Graphviz/DOT. This follows the same convention market-research-report, competitive-quadrant, trend-analysis, and c4-model-diagram already use in this suite. When a chart exceeds that range (more than roughly five segments, or a specific brand color palette), invoke the svg-charts skill instead (see below) rather than forcing it into a Mermaid block that would render illegibly.
• Hedge projections honestly: state the confidence basis for a projection (historical actuals, comparable-company benchmark, or a stated assumption) rather than presenting every number with uniform certainty.

Charts/graphics: Mermaid by default, svg-charts when it isn't enough

This genre needs revenue, market-segmentation, competitor, and use-of-funds charts, plus an optional org chart. Two things are true at once, both grounded in checked evidence rather than assumption:

• Mermaid covers most of it, and stays the default. Every diagram-bearing genre already in this suite (market-research-report, competitive-quadrant, trend-analysis, c4-model-diagram) embeds its chart guidance directly in its own SKILL.md as fenced mermaid blocks; business-plan does the same for anything within Mermaid's documented range (see the Rules section above).
• Mermaid has real, documented limits this genre will hit: pie-chart labels collide past roughly five slices, its base syntax has no per-segment custom-color override (colors come from a fixed theme), and it has no log-scale or multi-series combo support (https://www.mermaidcreator.com/blog/mermaid-pie-bar-chart-metrics-visualization, https://www.pkgpulse.com/guides/mermaid-vs-d3-vs-chartjs-diagrams-data-visualization-2026). A business plan with more than five named competitors, or one that needs to match a company's actual brand colors, will hit these limits in practice, not hypothetically.

For those cases this genre invokes the svg-charts helper skill, which generates a standalone .svg file (never inline markup — GitHub strips embedded <svg> elements from rendered markdown, so inline SVG would silently fail to render on the primary surface these documents are viewed on: https://alexwlchan.net/notes/2024/how-to-render-svgs-on-github/) and returns the <img src="..."> line to embed. See the svg-charts skill for its full decision rule and output contract.

Anti-triggers — do not use this genre for

• Scoping a single feature or product's build — that is prd (problem, goals, requirements for what to build) or feature-spec (the build-ready slice), not a whole-company plan for financing or strategy.
• A fieldwork-and-sampling study of a market at a point in time — that is market-research-report: disclosed sample frame, method, and fieldwork basis, not a company's own plan with financial projections and (optionally) a funding ask.
• A pure technical trade-off evaluation — that is engineering: a mandatory options-vs-criteria comparison table, not a business's operating and financial plan.
• A short, standalone decision brief for a reader with no time for the full plan — that is exec-summary. A business plan's own Executive Summary section serves this role within the plan; it is not a separate genre invocation.
• Tracking how a market is changing and projecting it forward — that is trend-analysis, not a single business's own operating plan.

MIF frontmatter

type: semantic — a business plan is declarative planning knowledge (what the business is, how it competes, its financial plan), not a time-bound event or a step sequence. Climb to L2 with namespace (business-plan/<area>), modified, title, and tags when the drafting context supplies them. Climb to L3 with provenance, citations[] tied to every sourced market/financial claim, and relationships[] (e.g. relates-to a companion market-research-report or prd for the same venture) when the plan is sourced from real, attributable input. Gate every output with mif-validate at its target level; the floor is --level 1.

Why machine-readable — the point of MIF here

The same document still reads as a fundable business plan and projects losslessly to JSON-LD and back — one artifact, two readers.

The L1 -> L3 climb (two exemplars)

This genre ships the same plan at two MIF levels so the climb is explicit:

• templates/good-l1.md — L1 floor: id, type, created + body. A complete, valid business plan, but opaque to a machine consumer.
• templates/good.md — L3 (highest this genre supports): adds namespace, modified, temporal validity, W3C-PROV provenance, citations[] tied to the market/financial claims, and a typed relationships[] graph.

Author at the highest level the drafting context supports (grade down rather than fabricate). templates/bad.md shows the antipattern: a plan that presents revenue projections and a funding ask with no disclosed assumptions anywhere — the genre's single most load-bearing requirement, missing.

Sources consulted for this genre's structure

1. U.S. Small Business Administration — Write your business plan, https://www.sba.gov/business-guide/plan-your-business/write-your-business-plan
2. SCORE — Business Plan Template for a Startup Business, https://www.score.org/resource/template/business-plan-template-a-startup-business
3. Ash Maurya, Running Lean / Lean Canvas — Business Model Toolbox, https://bmtoolbox.net/tools/lean-canvas/
4. SCORE — Financial Projections Template, https://www.score.org/resource/template/financial-projections-template
5. Guy Kawasaki — The Only 10 Slides You Need in Your Pitch (10/20/30 rule, investor-facing narrative discipline), https://guykawasaki.com/the-only-10-slides-you-need-in-your-pitch/