Investment Memo Generation

Investment Memo Generation

When to Activate

Trigger when user asks: "investment memo", "investment thesis", "comprehensive analysis of [company]", "research overview of [company]", "IC memo", "build a case for [company]", "write up [company]"

Memo Structure (7 Sections)

1. THESIS FRAMING: Core investment question, 3-5 thesis pillars, disconfirming evidence to monitor
2. FINANCIAL PROFILE: Revenue mix, growth rates, margin trends, Rule of 40 (for software), FCF generation
3. COMPETITIVE LANDSCAPE: Direct/indirect competitors, market position, competitive moat assessment
4. MANAGEMENT OUTLOOK: Forward guidance, strategic priorities, capital allocation
5. ANALYST PERSPECTIVES: Consensus views, price targets, rating distribution
6. RISK ASSESSMENT: Macro, regulatory, competitive, operational risks with leading indicators
7. CATALYSTS: 12-24 month bear/base/bull scenarios with probability weights

Labelling Convention

Label every paragraph as:

• [FACT]: Sourced directly from SEC filings, financial data, or official company statements
• [ANALYSIS]: Derived from quantitative comparison or trend analysis
• [INFERENCE]: Qualitative judgment or forward-looking assessment

Workflow

Step 1: Financial Foundation

Tool: segment_analyzer Query: Revenue by segment and geography, 3-year trends

Tool: fundamentals_analyzer + sec_financials Query: Key financial metrics, analyst estimates (high/mean/low), price targets, margins, EPS

Present:

• Company identification: "[Company] ([Ticker]) — [Sector]"
• Financial snapshot: Revenue, EPS, margins for latest period
• Analyst consensus: [X] analysts, average target $[X]

STOPPING POINT 1 — Financial Data Gathered

"I've pulled the financial foundation. Here's a quick snapshot:

• Revenue: $[X]B ([+/-X]% YoY)
• EPS: $[X] (consensus: $[X])
• Margins: Gross [X]%, Operating [X]%

I can:

• Continue to qualitative research (SEC filings, broker research, earnings calls, press releases)
• Adjust the scope (different time period, specific segments to focus on)
• Run earnings intelligence first (deep-dive recent quarter) → loads earnings-intelligence skill
• Check competitive positioning (peer comparison before thesis) → loads competitive-intelligence skill

How would you like to proceed?"

Step 2: Qualitative Research

Tool: search_sec_filings Query: Risk factors, MD&A narrative, competitive advantages from 10-K

Tool: search_external_docs (filter: broker_research) Query: Analyst views, competitive positioning, sector analysis

Tool: search_company_events Query: Management guidance, strategic commentary from earnings calls

Tool: search_external_docs (filter: press_releases) Query: Recent catalysts, corporate developments

STOPPING POINT 2 — All Research Gathered

"Research complete. Key findings:

• Thesis direction: [Bullish / Cautious / Mixed] based on data gathered
• Top catalyst: [1-sentence catalyst]
• Top risk: [1-sentence risk from SEC filings]
• Management tone: [Confident / Cautious / Defensive] based on earnings commentary

I can:

• Generate the full 7-section memo (recommended)
• Review findings before synthesis (I'll present raw data for each section)
• Focus the memo (emphasise specific sections — e.g., risk-heavy or catalyst-focused)
• Generate IC-ready quick thesis (1-page version with recommendation)

What format works best?"

Step 3: Memo Compilation

Synthesise all data into 7-section format with [FACT]/[ANALYSIS]/[INFERENCE] labelling throughout.

See SECTION_TEMPLATES.md for detailed section templates.

Step 4: Optional Outputs

• PDF Generation: If user requests, call pdf_generator with report_title = "[Company] — Investment Memo", document_audience = "internal"
• Audience Adaptation: If user specifies audience, adapt per branching below

Audience Branching

• IC Memo (Investment Committee — full detail): All 7 sections with [FACT]/[ANALYSIS]/[INFERENCE] labelling, 4-6 pages
• Quick Thesis (PM/CIO — action-oriented): Thesis Framing + Financial Profile + Catalysts only, 1-2 pages, recommendation-first
• Client Summary (Client/Prospect — educational): Thesis Framing + Financial Profile in plain language, no internal jargon, balanced tone, 2-3 pages

Output Template

# Investment Memo: [Company] ([Ticker])

**Recommendation**: [BUY / HOLD / SELL] | **Conviction**: [High / Medium / Low]
**Date**: [Date] | **Analyst**: AI Research Co-Pilot

---

## Executive Summary
- [FACT] [Key financial metric]
- [ANALYSIS] [Key trend observation]
- [INFERENCE] [Forward-looking thesis statement]
- Recommendation: [BUY/HOLD/SELL] with [X]% upside to $[target]

## 1. Thesis Framing
**Core question**: [The key investment question this memo answers]

| Thesis Pillar | Evidence | Confidence |
|---------------|----------|------------|
| [Pillar 1] | [FACT: data point] | [High/Medium/Low] |
| [Pillar 2] | [FACT: data point] | [High/Medium/Low] |

**Disconfirming evidence to monitor**: [What would invalidate this thesis]

## 2. Financial Profile
| Metric | FY[X-2] | FY[X-1] | FY[X] | Trend |
|--------|---------|---------|-------|-------|
| Revenue | [X] | [X] | [X] | [direction] |
| EPS | [X] | [X] | [X] | [direction] |

## 3-7. [Remaining sections follow structure above]

Cross-Skill References

• "Deep-dive recent earnings?" → earnings-intelligence skill
• "Compare to competitors?" → competitive-intelligence skill
• "Check insider/institutional signals?" → insider-institutional-analysis skill
• "Generate PDF?" → pdf-report-generation skill
• "Adapt for different audience?" → audience-adaptive-narrative skill

Stopping Points

• After Step 1 (financial data): Offer scope adjustment + cross-skill options
• After Step 2 (all research gathered): Offer memo generation + format options
• After Step 3 (memo compiled): Offer PDF generation + audience adaptation