business-model

# Knowledge Frameworks — {SKILL_NAME}

This file augments the skill methodology with practitioner frameworks, case precedents,
and sector-specific analysis rubrics from the agentii knowledge base (spec 037).
It is progressively disclosed — loaded on demand, not at skill activation (FR-014).

## How to Use

Before executing the analysis, scan the relevant frameworks below for the target sector
and analysis type. Where a framework applies, incorporate its methodology alongside the
standard protocol. Cite knowledge entries by their `/v/knowledge/{citation_id}` link.

## Practitioner Frameworks

### Fundamental Analysis (L2)
- **Fisher Scuttlebutt Method**: Channel checks, supplier interviews, customer surveys — validate management claims through primary research.
- **Morningstar Moats**: Network effects, switching costs, intangible assets, cost advantage, efficient scale — score competitive durability.
- **Greenblatt Magic Formula**: Rank on earnings yield + ROIC — quantitative screen for quality-at-a-reasonable-price.
- **Mauboussin Expectations Investing**: Price-implied expectations vs. fundamentals — identify the gap.

### Management Assessment
- **CEO/CFO track record**: Past capital allocation decisions, M&A history, share buyback timing.
- **Incentive alignment**: Compensation structure, insider ownership, option grants.

### Sector-Specific Playbooks
- **Healthcare (Pharma/Biotech)**: Pipeline depth analysis, FDA catalyst calendar, patent cliff exposure, drug pricing risk.
- **Technology**: Adoption S-curves, network effects quantification, R&D efficiency, platform ecosystem lock-in.
- **Financials**: Spread analysis (NIM), credit cycle positioning, loan loss reserve adequacy, capital return capacity.
- **Consumer**: Brand strength metrics, unit economics, channel mix trends, customer acquisition cost payback.

## Historical Analogues

When the current situation matches a known pattern, query `search_by_analogue` with the
appropriate `market_regime`/`event_type`/`company_situation` tags. Cite relevant cases as
precedent.

## References

All frameworks sourced from `gold.knowledge_entries` (spec 037). See `contracts/citation-and-memory.md`
for citation format conventions.

# business-model — Methodology Detail

Extracted from SKILL.md for progressive disclosure (US5).

## Retrieval Strategy

Follow the retrieval strategy decision tree in `contracts/retrieval.md`:

- **Branch (a)** — `search_xbrl_facts` for revenue concentration, gross margin profile, segment-level P&L (concepts `Revenues`, `GrossProfit`, `OperatingIncomeLoss`, `SegmentReportingInformation`).
- **Branch (b)** — `search_cross_period` when analyzing channel-mix evolution over 2+ years (mode 1_2) or management-change continuity (mode 1_5).
- **Branch (c)** — single-period three-layer for the most recent 10-K/10-Q business-overview pages (modes 1_1, 1_3, 1_4).

**Layer 1 narrowing — `secondary_labels` filter **: For 8-K-driven business-model-relevant disclosures, prefer `?secondary_label=other_events_8_01` (item 8.01 covers business-strategy events) AND `?secondary_label=financial_results_2_02` (item 2.02 surfaces segment-level revenue commentary). The GIN-indexed filter on `pipeline.src_documents.secondary_labels` is the preferred narrowing axis BEFORE Layer 2.

**Layer 2 page-relevance signal**: Score pages using `labels->>'general'->>'description'` (~100-char LLM-generated page summary) AND `labels->>'general'->>'keywords'` (extracted entity terms). Both fields are populated on 96%+ of `pipeline.src_silver_pages` rows. For business-model analysis, prefer pages whose `keywords` contain entity terms like product names, segment names, geographies, channel partners, executive names.

## Protocol

1. ** Pre-flight (mandatory)**: call `get_company_fiscal_calendar/{ticker}` for fiscal orientation, then `get_ticker_coverage/{ticker}` to discover which data sources are populated. Route based on coverage: `sec_filings` populated → standard three-layer protocol; `xbrl_facts`-only → structural inferences from segment data only, flag `data_availability: degraded`.

2. **XBRL retrieval FIRST (grounding truth)**: `search_xbrl_facts(ticker, concept=["Revenues","GrossProfit","OperatingIncomeLoss","SegmentReportingInformation"], fiscal_year=[<latest>])` for segment/product/geography/channel revenue & margin breakdowns BEFORE any document discovery — returns `is_primary: true` rows by default (superseded rows hidden; `?include_all_sources=true` only for audit). The `source_authority` field (3=10-K, 2=10-Q, 1=8-K) is returned for fact-provenance transparency. This is the structural backbone of the analysis.

3. **Layer 1 Document Discovery**: `search_documents(ticker={T}, form_type=["10-K","10-Q","20-F"], limit=3)` to find the most recent annual report (richest business-overview content) plus trailing 10-Qs. For foreign issuers route to `form_type=["20-F","6-K"]`. Add `?secondary_labels=financial_results_2_02,other_events_8_01` to also surface relevant 8-Ks. Document identifiers returned in `citation_id` form (`sec135` for SEC filings, `ect83` for earnings call transcripts) — pass these directly to Layer 2.

4. **Layer 2 Page Map**: `read_source_outline/{ticker}/{citation_id}` — scan `description` + `keywords` for each page. Identify pages covering: Business overview / Item 1 (mode 1_1), Distribution & sales channels (mode 1_2), Revenue by segment & geography (mode 1_3), Industry / market context (mode 1_4), Directors & executive officers (mode 1_5). Bare `page_no` integers are forbidden in LLM-facing output — always use `{ticker} {citation_id} page<N>` format (e.g., "LLY sec135 page12").

5. **Layer 3 Deep Read**: `read_source_pages/{ticker}/{citation_id}?pages=page<N1>,page<N2>` — load full `page_content` for ONLY the 3-5 pages identified in Layer 2 to enrich the XBRL backbone with narrative.

6. **Cross-period (when applicable)**: use `search_cross_period` / `batch_search` for channel-mix evolution (mode 1_2) or management-change continuity (mode 1_5).

7. **Evidence-pack handoff**: produce `evidence-pack.json` + `evidence-digest.md` . All citations use the v1.0 frozen format with `{ticker} {citation_id} page<N>` references.

## Production Grounding

The Neon production database and `api.agentii.ai` REST/MCP surfaces are LIVE and AUTHORITATIVE as of 2026-05-25. All retrieval planning MUST treat these as ground truth. Production scale: 15.99M `gold.xbrl_facts` (with `is_primary` partial index), 51,089 `pipeline.src_documents` (100% non-null `description`, GIN-indexed `secondary_labels`), 1.34M `pipeline.src_silver_pages` (covering ALL 5 SEC form types — 8-K/10-K/10-Q/6-K/20-F; `labels` JSONB with `general.description` + `general.keywords`), 75,967 `pipeline.earnings_calendar` rows, 142 `gold.launch_ticker_registry` tickers at 100% processing. Always call `get_ticker_coverage/{ticker}` before retrieval planning.

# Measuring the Moat — Institutional Methodology

Methodology synthesized from institutional investment research; all text is an original paraphrase.

---

## Sustainable Value Creation: The Dual-Dimension Framework

Sustainable value creation has two dimensions that must be assessed together. Neither alone is sufficient:

1. **Magnitude**: How large is the spread between return on invested capital (ROIC) and the weighted average cost of capital (WACC)? This measures the economic profit generated per dollar of invested capital today.

2. **Sustainability**: How long can the company maintain this spread? This is the Competitive Advantage Period (CAP) — the duration over which the company can find and invest in opportunities that earn returns above the cost of capital.

A company with a wide current spread but a short CAP (e.g., a pharmaceutical company approaching a patent cliff) may be worth less than a company with a narrower but highly persistent spread (e.g., a consumer staples franchise).

### The Moat Formula

```
Sustainable Value Creation = (ROIC − WACC) × Investment Capacity × CAP Duration
```

- **ROIC − WACC spread**: Magnitude of value creation per dollar invested
- **Investment Capacity**: How much capital the company can deploy at this spread
- **CAP Duration**: How many years the spread can be maintained

---

## Why Sustainable Value Creation Is Hard

Creating value for a long time is extraordinarily difficult. Three forces work against it:

### 1. Competition Is a Magnet
Companies that invest a lot at high returns attract competitors. Empirical evidence shows that high returns on investment and rapid investment growth are associated with sharp subsequent declines in returns. The very success that attracts investors also attracts the competition that will erode it.

### 2. Market Maturation Eliminates Tailwinds
When end markets mature, competitors no longer benefit from market expansion. Growth-seeking shifts from expanding the market to gaining market share — a zero-sum contest that compresses returns for all participants.

### 3. Best Practices Diffuse
Operational excellence, once achieved, tends to spread across an industry. Supply chain innovations, pricing algorithms, customer acquisition techniques — all diffuse over time, dissipating the differences between firms.

### 4. Luck Regresses
A significant portion of corporate results is attributable to factors outside management's control. When luck plays a large role in short-term outcomes, regression toward the mean is rapid. Outcomes far from the average are followed by outcomes closer to the average.

---

## ROIC Regression Toward the Mean: The Empirical Evidence

ROIC is not stationary. Companies with high ROIC today tend to have lower ROIC tomorrow. The speed of regression varies dramatically by industry.

### Sector-Level ROIC Autocorrelation (5-Year Persistence)

| Sector | 5-Year ROIC Autocorrelation | Interpretation |
|--------|:---:|------|
| Consumer Staples | r = 0.46 | Highest persistence — brands, distribution networks, and switching costs create durable moats |
| Healthcare | r = 0.38 | High persistence — patents, regulatory barriers, and specialized expertise slow competition |
| Technology | r = 0.28 | Moderate persistence — network effects and switching costs provide some durability, but disruption risk is elevated |
| Industrials | r = 0.22 | Below-average persistence — operational efficiency advantages erode over time |
| Financials | r = 0.19 | Low persistence — financial innovation and regulatory changes rapidly compress spreads |
| Energy | r = 0.15 | Lowest persistence — commodity prices drive ROIC; competitive advantages are minimal |

**Key Insight**: A company in the Energy sector with a current ROIC of 25% is FAR less likely to sustain that return than a Consumer Staples company with the same 25% ROIC. The sector-level autocorrelation calibrates the sustainability assessment. A company in an industry with r = 0.15 should be assumed to mean-revert rapidly regardless of current performance.

### Quintile Transition Data (2013-2023, Russell 3000)

Over a decade, ROIC quintiles reverted substantially. Companies in the top quintile in 2013 saw their median ROIC decline by approximately 8-10 percentage points relative to the universe median by 2023. Companies in the bottom quintile saw improvements of 5-8 percentage points.

The practical implication: when a company's current ROIC is an extreme outlier (top or bottom quintile), the base rate strongly suggests regression toward the mean. The analyst's job is to identify the specific factors that would make THIS company an exception to the base rate.

---

## Industry Structure Classification

Industry structure determines the competitive dynamics that govern ROIC persistence. Classify the industry before assessing the individual company.

### Five Classification Types

| Structure | Characteristics | ROIC Persistence | Examples |
|-----------|---------------|:---:|----------|
| **Fragmented** | Many competitors, low concentration, no dominant player. Price competition common. | **Low** — excess returns competed away rapidly | Restaurants, local services, residential construction |
| **Oligopoly** | Few large competitors, high concentration. Implicit or explicit coordination on pricing. | **Moderate-High** — pricing discipline can sustain returns | Aerospace (Boeing/Airbus), credit rating agencies, soft drinks |
| **Dominant Firm** | One firm with significant market share and cost/scale advantages. Followers compete on niche dimensions. | **High for dominant firm**; low for followers | Search engines, enterprise operating systems, warehouse retail |
| **Network/Rising Returns** | Each additional user/customer increases the value of the platform. Winner-takes-most dynamics. | **Very High for the winner**; negative for others | Social networks, online marketplaces, payment networks |
| **Commodity** | Undifferentiated product. Price is the sole competitive dimension. Cost position is everything. | **Very Low** — no pricing power | Bulk chemicals, agricultural commodities, shipping |

### Testing the Classification

Industry structure is not self-evident. Test with quantitative evidence:

1. **Market Share Stability**: Calculate the 5-year instability index (sum of absolute market share changes / number of firms). High instability → Fragmented or Commodity. Low instability → Oligopoly or Dominant Firm.

2. **Concentration Ratio**: CR4 (top 4 firms' market share). > 60% → Oligopoly or Dominant Firm. < 30% → Fragmented.

3. **Pricing Power**: Gross margin stability over 5+ years. Stable or expanding → barriers exist. Continuously contracting → Commodity or Fragmenting.

---

## Five Forces + Barriers Integration

### Industry Analysis Framework

1. **Industry Map**: Map the full value chain from suppliers to end customers. Identify where power is concentrated and where margins pool.

2. **Profit Pool**: Quantify the total profit generated at each stage of the value chain. Where is the money actually being made? The largest revenue stage may not be the most profitable stage.

3. **Market Share Instability**: Measure how much market share shifts year-to-year. High instability = low barriers.

4. **Concentration Trends**: Is the industry consolidating or fragmenting? Consolidation typically increases pricing power; fragmentation reduces it.

### Barriers That Protect Incumbents

| Barrier Type | Description | Durability |
|-------------|------------|:---:|
| **Switching Costs** | Customer cost (time, money, risk) of changing providers | **High** — structural, self-reinforcing |
| **Network Effects** | Platform value increases with user count | **Very High** — winner-takes-most dynamics |
| **Intangible Assets** | Brands, patents, regulatory licenses | **Moderate-High** — patents expire, brands can erode |
| **Cost Advantages** | Scale economies, proprietary technology, favorable locations | **Moderate** — can be replicated with sufficient investment |
| **Distribution Control** | Exclusive or semi-exclusive access to distribution channels | **Moderate** — channels evolve over time |
| **Regulatory Protection** | Government-granted monopolies or oligopolies | **Variable** — regulation can change |

---

## Disruption & Dis-Integration Analysis

### When Moats Fail

Moats are not permanent. Two primary mechanisms cause moats to fail:

**Disruptive Innovation**: A new entrant serves an overlooked segment with a simpler, cheaper, or more accessible product. The incumbent, focused on its most profitable customers, dismisses the threat — until the entrant moves upmarket and captures the core.

**Industry Dis-Integration**: Vertically integrated industries can dis-integrate when modular interfaces standardize. The value shifts from the integrated system to specific layers. Personal computers dis-integrated from vertically integrated systems (Apple II era) to modular components (Wintel era). The moat shifted from the assembler to the component suppliers with the strongest positions.

### Testing for Disruption Risk

1. Is there a segment of the market that incumbents are happy to ignore or actually prefer to lose?
2. Is there a technology or business model that enables a lower-cost or more accessible alternative?
3. Can the disruptor improve its offering over time to attract the incumbent's core customers?

If all three answers are yes, the moat is at risk regardless of current ROIC.

---

## Moat Assessment Checklist

### Category 1: Industry Structure (15 items)
- [ ] Industry concentration (CR4) quantified
- [ ] Market share stability over 5 years assessed
- [ ] Industry structure classified (Fragmented/Oligopoly/Dominant/Network/Commodity)
- [ ] Profit pool mapped across value chain
- [ ] Barriers to entry identified and classified
- [ ] Threat of substitutes assessed
- [ ] Buyer concentration and bargaining power assessed
- [ ] Supplier concentration and bargaining power assessed
- [ ] Regulatory barriers identified (licenses, permits, approvals)
- [ ] Capital intensity of entry assessed (minimum efficient scale)
- [ ] Industry growth rate (expanding markets mask competitive intensity)
- [ ] Industry ROIC dispersion (wide = differentiation possible; narrow = commodity)
- [ ] Exit barriers (high exit barriers → prolonged overcapacity)
- [ ] Technology disruption risk assessed
- [ ] Business model disruption risk assessed

### Category 2: Company Position (18 items)
- [ ] Market share trend (gaining, stable, losing)
- [ ] Relative market share (company share / largest competitor share)
- [ ] Pricing power evidence (gross margin stability, ability to pass through costs)
- [ ] Customer retention / churn rate (if available)
- [ ] Net Promoter Score or customer satisfaction trend
- [ ] Brand strength (pricing premium vs. generic/competitor)
- [ ] Switching cost magnitude for customers
- [ ] Network effect strength (if applicable)
- [ ] Cost position relative to competitors (cost leader, parity, or disadvantage)
- [ ] Scale advantage quantification (volume relative to MES)
- [ ] Proprietary technology or process advantages
- [ ] Patent portfolio strength and expiry calendar
- [ ] Distribution advantages (channel exclusivity, shelf space, geographic coverage)
- [ ] Customer concentration (revenue from top 3 customers)
- [ ] Supplier concentration and dependency
- [ ] Employee retention and talent moat
- [ ] Data advantages (proprietary datasets that improve with scale)
- [ ] Regulatory or legal moats (licenses, spectrum rights, certificates of need)

### Category 3: Financial Evidence (17 items)
- [ ] 5-year ROIC trend (stable, improving, declining)
- [ ] ROIC − WACC spread magnitude and trend
- [ ] Sector ROIC autocorrelation applied (how fast does ROIC mean-revert here?)
- [ ] NOPAT margin trend (pricing power or cost pressure?)
- [ ] Invested capital turnover trend (capital efficiency improving or declining?)
- [ ] Gross margin stability (coefficient of variation over 5 years)
- [ ] Operating margin trend (operating leverage or cost creep?)
- [ ] FCF conversion ratio (FCF/NI > 70%?)
- [ ] ROIIC over rolling 3-year periods (positive and above WACC?)
- [ ] Revenue growth consistency (standard deviation of annual growth rates)
- [ ] Market share gain/loss correlation with pricing changes
- [ ] SBC as % of revenue (excessive dilution undermines moat claims)
- [ ] R&D efficiency (revenue per dollar of R&D spend)
- [ ] S&M efficiency (revenue per dollar of S&M spend)
- [ ] Working capital efficiency trend
- [ ] Capex/Revenue ratio and trend
- [ ] Dividend and buyback sustainability (are returns of capital consistent with reinvestment needs?)

### Category 4: Management & Governance (10 items)
- [ ] CEO tenure and track record
- [ ] Management depth (Alpha/Beta/Delta assessment)
- [ ] Capital allocation track record (M&A returns, buyback timing, dividend policy)
- [ ] Insider ownership alignment
- [ ] Compensation tied to ROIC or returns-based metrics (not just revenue or EPS)
- [ ] Board independence and expertise
- [ ] Strategy consistency (frequent pivots → no durable moat)
- [ ] Management communication candor about competitive challenges
- [ ] Succession planning visibility
- [ ] Corporate culture indicators (employee satisfaction, Glassdoor ratings, turnover)

### Category 5: Forward Indicators (7 items)
- [ ] Competitive announcements (new entrants, capacity additions, price wars)
- [ ] Technology trends affecting industry structure
- [ ] Regulatory changes pending
- [ ] Customer behavior shifts (channel migration, demographic changes)
- [ ] Supplier industry consolidation or fragmentation
- [ ] Patent cliff calendar (for IP-dependent industries)
- [ ] Macro sensitivity of competitive position (does the moat strengthen or weaken in downturns?)

### Scoring

| Score Range | Assessment | Action |
|:---:|------|------|
| 55-67 | Strong, durable moat | Proceed with confidence; focus on valuation |
| 40-54 | Moderate moat | Proceed with caution; monitor disruption risk |
| 25-39 | Weak or eroding moat | High risk for long positions; potential short candidate |
| < 25 | No moat / commodity | Avoid longs unless deep value with specific catalyst |

---

## The Role of Government

Government action can create, protect, or destroy moats:

- **Create**: Patents, regulatory licenses, spectrum auctions, certificates of need
- **Protect**: Antitrust enforcement (or lack thereof), tariffs, domestic content requirements
- **Destroy**: Antitrust breakup, deregulation, compulsory licensing, price controls

Government-created moats are inherently less durable than market-created moats because they depend on political will, which can shift with elections.

---

## Brands as Moats

Brands are among the most commonly claimed — and most commonly misidentified — moats. A brand is a moat ONLY when it changes customer behavior:

- **Pricing Power Brand**: Customers pay a premium for the branded product vs. the generic equivalent. Coca-Cola, Louis Vuitton, Hermès.
- **Trust Brand**: Customers choose the branded product to reduce risk. Moody's, S&P, UL certification.
- **Habit Brand**: Customers choose out of routine, not preference. Most CPG products. NOT a moat — habits can change.

A brand that customers recognize but won't pay more for is a marketing asset, not a competitive advantage.

# business-model — Analyst Mode Definitions

Extracted from SKILL.md for progressive disclosure (US5). The skill body keeps a pointer under `## Methodology → Analyst Modes`.

### Mode: business-model-classification (1_1 — anchor)

**Display name**: Business Model & Offerings Classification

<!-- ported_from: references/prompts/1/1_1.md -->

**Objective**: Determine business-model type (product / service / platform), core offering, and market positioning (low-end / mid-tier / high-end) using the most recent 10-K (annual) and trailing 10-Q/20-F disclosures, supplemented by web search only where XBRL/filing data is insufficient.

**Output structure**:

- **Business Model**: [Product / Service / Platform]
- **Core Offering**: [e.g., Connected Wearable, Diagnostic Consumables, SaaS Subscription, Drug Pipeline, Cloud Platform]
- **Positioning**: [Low-end / Mid-tier / High-end] + brief rationale (e.g., "High-end based on >70% gross margin and premium ARPU vs. peers")
- **Citation density**: ≥1 citation per 200 words, format `{ticker} {citation_id} page<N>`.

### Mode: distribution-channel-analysis (1_2)

**Display name**: Distribution Channels & Go-to-Market Analysis

<!-- ported_from: references/prompts/1/1_2.md -->

**Objective**: Assess primary distribution model (direct sales / channel partners / hybrid), channel mix evolution (3-year trailing), and strategic implications for pricing power and customer intimacy.

**Output structure**:

- **Distribution Model**: [Direct Sales / Channel Partners / Hybrid]
- **Distribution Partners**: [list disclosed channel types and representative partners]
- **Current Channel Mix**: Direct : Indirect = 1 : XX (latest available data)
- **Historical Channel Mix Trend (Trailing 3 Years)**:
 - Year -2: [Direct : Indirect = 1 : XX]
 - Year -1: [Direct : Indirect = 1 : XX]
 - Current: [Direct : Indirect = 1 : XX]
- **Strategic Implication**: e.g., "Shift toward direct sales has enhanced pricing control and customer intimacy but increased SG&A".

### Mode: revenue-composition-and-concentration (1_3)

**Display name**: Revenue Composition & Concentration Risk Analysis

<!-- ported_from: references/prompts/1/1_3.md -->

**Objective**: Decompose revenue by product line / customer type / geography / end market, identify concentration risk (any single product or client >20% of total revenue), and trace temporal mix evolution.

**Output structure**:

- **Latest annual + trailing-quarter revenue breakdown (XBRL segment data)**:
 - By Product Line: top 3 contributors with %
 - By User Type (B2B vs. B2C): mix with %
 - By Geography (NA / EMEA / APAC): top 3 regions with %
 - By End Market: top 3 markets with %
- **Concentration Risk Matrix**: any product/client >20% flagged.
- **Temporal Comparison**: vs. prior quarter and YoY same-period.

### Mode: market-sizing-and-relative-growth (1_4)

**Display name**: Market Sizing (TAM/SAM/SOM) & Relative Growth

<!-- ported_from: references/prompts/1/1_4.md -->

**Objective**: Quantify TAM, SAM, SOM for key verticals, project 3-5 year CAGR, assess relative growth (company vs. addressable market) and historical SOM evolution.

**Output structure**:

- **Market Sizing & Growth**:
 - TAM (Current Year): USD XXX bn
 - SAM (Current Year): USD XXX bn
 - SOM (Current Year): XX%
 - TAM CAGR (Past 3 Years): XX%
 - TAM CAGR (Forward 3-5 Years): XX%
- **Relative Growth Table**: Company Revenue Growth vs. TAM CAGR for past 3Y and next 3Y.
- **Historical SOM Trajectory**: 3-year evolution with execution-strength assessment.

### Mode: management-and-leadership (1_5)

**Display name**: Management Team & Leadership Analysis

<!-- ported_from: references/prompts/1/1_5.md -->

**Objective**: Assess executive team backgrounds, track records, and recent leadership changes (CEO/CFO/COO/CMO) for strategic-execution implications.

**Output structure**:

- **Key Executives & Track Record**: [Name, Role, Tenure, Notable Prior Experience, Industry Expertise, Capital Allocation Record]
- **Recent Management Changes (Trailing 1-2 Quarters)**: [Name, Role, Effective Date, Reason, Successor Background]
- **Strategic Implications of Changes**: e.g., "New CFO brings strong M&A background, suggesting a shift toward inorganic growth".

# business-model — Output Structure (full template)

Extracted from SKILL.md for progressive disclosure (US5). The skill body keeps a compact summary under `## Output Structure`.

The final deliverable MUST be written as a markdown file to the workspace using the convention :

```
{ticker}/{YYYY-MM-DD_HHMM}_business-model_{affix}.md
```

Where `affix` is a short descriptive slug (e.g., `product-line-decomp`, `channel-mix`, `market-sizing`, `management-changes`). Examples:

- `LLY/2026-05-25_1430_business-model_product-line-decomp.md`
- `NVDA/2026-05-25_1545_business-model_platform-classification.md`

The deliverable file MUST contain (in order):

1. **Executive Summary** (≤200 words) — business-model classification + headline structural insights.
2. **Business Model Type** (mode 1_1) — Product / Service / Platform classification with rationale.
3. **Product Line Decomposition** (mode 1_3) — revenue breakdown by product, top-3 contributors, concentration risk.
4. **Distribution Channel Analysis** (mode 1_2) — direct vs. indirect mix, partners, 3-year evolution.
5. **Customer Segment Analysis** (mode 1_3) — B2B vs. B2C, geography, end markets, concentration risk.
6. **Revenue Model** — recurring vs. one-time, pricing power, unit economics, gross-margin profile.
7. **Business Unit Performance** — segment-level P&L where available (XBRL or narrative).
8. **Market Sizing & Competitive Positioning** (mode 1_4) — TAM/SAM/SOM, relative growth vs. market.
9. **Management & Leadership** (mode 1_5) — executive team, recent changes, strategic implications.
10. **Coverage Gaps & Citations** — list of dimensions not retrievable + full citation index in `{ticker} {citation_id} page<N>` format.

**Citations & memory**: follow `contracts/citation-and-memory.md` — ≥1 citation per 200 words; every material fact, table row, and metric is immediately followed by its inline clickable `https://agentii.ai/v/{ticker}/{citation_id}/{N}` link; a bottom **Citations** section provides a non-duplicative roll-up index; the closing TUI reply includes a compact **Key Citations** list (headline 5–10 facts) of clickable `/v/` URLs; and append the run to `agentii.md` per `contracts/agentii-md-schema.md`.

# business-model — Tool Fallbacks

Extracted from SKILL.md for progressive disclosure (US5).

| Tool | Failure Mode | Fallback Action | Coverage Annotation |
|------|-------------|-----------------|---------------------|
| `read_source_pages` | SQL error / PROXY_ERROR | Use `search_keyword_in_source(document_id, keyword)` if document_id known; otherwise `search_documents` with same query | "source file unavailable; used keyword search instead" |
| `read_source_deep_outline` | PROXY_ERROR / 404 | Use lightweight `read_source_outline` and flag `deep_outline_degraded: true` | "deep outline unavailable; used lightweight page map instead" |
| `read_source_outline` | PROXY_ERROR / 404 | Use `list_sources` for document-level metadata | "page map unavailable; used document listing instead" |
| `list_xbrl_concepts` | Timeout / 503 | Use direct `search_xbrl_facts` with standard US-GAAP concepts (Revenues, GrossProfit, OperatingIncomeLoss) | "concept discovery skipped due to timeout; using standard US-GAAP concepts" |
| `get_company_fiscal_calendar` | Cross-validation failed | Use XBRL-derived period grid from `search_xbrl_facts` `period_end` dates | "fiscal calendar mismatch; using XBRL-derived period grid" |
| `search_xbrl_facts` | Empty (concept not populated) | Fall back to narrative analysis from 10-K Item 1 (Business Overview) | "segment-level P&L not in XBRL; structural inference only" |
| `batch_search` | PROXY_ERROR | Use sequential individual calls (one per sub-query) | "batch search unavailable; used sequential calls" |

Tool errors are retried ONCE with the fallback action before escalating to the retrieval gaps failure policy. If both Layer 2 and Layer 3 tools are unavailable, enter document access degradation mode (structured data + metadata only, flag output as degraded).

# business-model -- Buy-Side Stock Analysis Methodology

Institutional-grade framework for fundamental equity research. Applied to qualitative company analysis, business model evaluation, and competitive positioning.

## Protocol

### 1. Idea Generation

Three channels:
- **Quantitative screens**: Combine a valuation metric with an operating metric (e.g., revenue $100M-$500M, EBITDA margin above 35%, EV/EBITDA below 5x). Customize to fund mandate. Screens are a starting point, not the sole source.
- **Follow respected investors**: Review 13D filings, investor letters, and public portfolios of top performers.
- **Continuous reading**: Newspapers, industry journals, financial websites, casual observation of consumer behavior. Maintain a critical eye for dislocations between market perception and operational reality.

### 2. Establish the Investment Thesis

Articulate precisely why the stock warrants analysis. Identify catalysts: activist involvement, contrarian entry at 52-week lows, embedded asset value, respected holders signaling conviction. The thesis must be falsifiable.

### 3. Two-Part Business Analysis

**Part A -- Historical and current state**: What does the company sell and how does it generate cash flow? What is the stated strategy, and does the transaction and organic growth record support it? How has the company performed against its own narrative?

**Part B -- Forward outlook**: Are cash flows sustainable given capital intensity and competitive dynamics? What are the specific risks, and where can deterioration originate? Segment by segment, distinguish organic from acquisition-driven growth.

### 4. Industry Analysis

Companies cannot be analyzed in isolation:
1. Begin with public filings, extend to competitor filings for contrasting perspectives.
2. Consult industry-specific trade associations for structural data.
3. Use equity research and market research for sector overview.
4. Apply established frameworks as thinking prompts, not boundary limits.
5. Analyze demand drivers (demographics, structural tailwinds), supply dynamics (construction pipeline, local market cycles, inventory concentration versus company footprint), and industry structure (fragmentation, consolidation runway, barriers to entry).
6. Independently verify management claims about supply and competitive dynamics. Management's definition of terms may not match analytical requirements.

### 5. Management Team Assessment

Evaluate whether management has the background and incentives to execute the stated strategy:
- **Three-axis experience matrix**: Industry experience, industry-specific operations experience, and general operating experience. Flag C-suite leaders lacking operational backgrounds, especially for acquisition-heavy strategies where integration execution is critical.
- **Turnover analysis**: Is change positive (required shakeup after underperformance or shareholder pressure) or negative (instability)?
- **Compensation alignment**: Deconstruct stock awards (RSUs versus options, time-based versus performance-based). The critical test: is compensation tied to share price, or to metrics that may not align with shareholder value? If the thesis requires a specific catalyst, confirm management is incentivized to pursue it.

### 6. Board of Directors Evaluation

Map each director across four expertise axes: industry experience, asset-class experience, public company governance, and capital markets. Note committee assignments, tenure, internal versus external status, and activist-appointed directors as potential catalysts. Board composition is particularly important when the thesis depends on a specific value-unlocking event.

### 7. Primary Research

When secondary sources are insufficient, primary research creates the analytical edge:
- Draw up an extensive question list, divided into "must-have" and "nice-to-have" tiers.
- Conduct site visits. Observe operations directly.
- Triangulate findings against management's public narrative.

### 8. Trust but Verify and Position Building

Every management claim must be independently pressure-tested. Conviction is built incrementally: enter with a small position while continuing due diligence, using identified downside cushions (embedded asset value, hard catalysts) as margin of safety. Scale into the position as evidence accumulates; exit if the thesis deteriorates. This contrasts with binary Buy/Sell ratings and enables entry before full conviction is achieved.