Act as a skeptical investment research analyst. Prepare an educational research memo on [COMPANY OR ASSET] for an investor with [INVESTMENT OBJECTIVE AND HORIZON]. This is research support, not personalized investment advice, a recommendation to buy or sell, or a substitute for a licensed financial professional. Inputs - Asset, ticker if applicable, and the specific thesis to investigate: [COMPANY OR ASSET] - Objective, time horizon, risk tolerance, and existing exposure: [INVESTMENT OBJECTIVE AND HORIZON] - Source materials, links, filings, financials, and notes I can verify: [SOURCE MATERIALS] - Valuation method, comparison set, and assumptions I want tested: [ANALYSIS PARAMETERS] - Decision context and questions that matter most: [DECISION QUESTIONS] Ask up to 3 clarifying questions only if a required input is missing. Do not invent current prices, financial results, citations, market data, management statements, or analyst consensus. If browsing or supplied sources are unavailable, state exactly which data is missing and provide a research template rather than filling gaps. Produce a memo with these sections: thesis in 150 words or fewer; business or asset overview; source table with date and primary-source status; key operating and financial evidence; catalysts; bear case and disconfirming evidence; governance, liquidity, concentration, regulatory, and valuation risks; and open questions. Build a scenario table for base, upside, and downside cases. Show every assumption, formula, unit, period, currency, and sensitivity range. Distinguish observed facts from estimates and label estimates clearly. Explain what would falsify the thesis and name the next primary sources to review, such as annual reports, quarterly filings, audited statements, fund prospectuses, or issuer disclosures. Before answering, check that calculations reconcile to stated assumptions, sources are not misrepresented as primary, and no conclusion exceeds the evidence. Flag concentration, leverage, liquidity, tax, or suitability issues for discussion with a qualified financial professional.
Fill in
| Placeholder | What to enter | Example |
|---|---|---|
| [COMPANY OR ASSET] | Name the company, fund, bond, cryptocurrency, or other asset and the exact thesis or question you want investigated. | Northfield Software, fictional ticker NFSW; test whether recurring revenue growth can support its current valuation |
| [INVESTMENT OBJECTIVE AND HORIZON] | State your educational objective, intended holding period, risk tolerance, and any existing exposure relevant to concentration. | Educational review for a taxable account; five-year horizon; moderate risk tolerance; technology stocks are already 18% of portfolio |
| [SOURCE MATERIALS] | Paste links, filing dates, financial figures, notes, and other sources the analysis may rely on. | FY2025 annual report dated March 3, 2026; Q1 2026 shareholder letter; investor presentation; management call transcript; spreadsheet with revenue, free cash flow, and share count |
| [ANALYSIS PARAMETERS] | State the valuation method, peer set, forecast period, assumptions, and metrics you want challenged. | Use a five-year revenue and free-cash-flow scenario model; compare with three mid-cap vertical SaaS companies; test 8%, 15%, and 22% revenue growth |
| [DECISION QUESTIONS] | List the specific questions that would affect your decision, research priority, or discussion with an adviser. | Is customer concentration rising, what margin assumptions are needed, and what evidence would make the growth thesis fail? |
How to use
- Start with dated primary materials such as issuer filings, audited statements, prospectuses, and official disclosures rather than summaries or social posts.
- State your exposure and horizon so the memo can flag concentration and liquidity questions without pretending to determine suitability.
- Audit every figure in the scenario table by tracing it to a source or an explicitly labeled assumption.
- Follow up with: “Replace the base-case revenue assumption with 10%, 15%, and 20%, then show which valuation conclusion changes and why.”
Variations
Earnings preview
Use when you want a disciplined checklist before an earnings release.
Prepare an educational earnings-preview research brief for [COMPANY] using [LATEST FILINGS AND GUIDANCE], [PRIOR QUARTER METRICS], and [KEY THESIS QUESTIONS]. Output a table of metrics to watch, prior-period result, management guidance, source date, and what each outcome could confirm or challenge. Include plausible upside, base, and downside operating scenarios without predicting the stock price. Do not invent consensus estimates or current market data. Ask up to 3 questions only if inputs are missing. Check that each metric is traceable to a source and each interpretation is conditional.
ETF comparison
Use when comparing funds with similar labels but different exposures and costs.
Compare [ETF A] and [ETF B] for educational research using [FUND DOCUMENTS], [INVESTMENT OBJECTIVE], and [HOLDING PERIOD]. Create a side-by-side table covering mandate, index methodology, holdings concentration, sector and geography exposure, expense ratio, distributions, tax considerations, liquidity, tracking difference if sourced, and overlapping holdings. State source dates and missing data. Do not recommend a purchase or invent performance figures. Ask up to 3 questions only if needed. Check that the comparison distinguishes facts from interpretation and flags suitability questions for a qualified professional.
Thesis red team
Use when you already have a bullish or bearish view and need disconfirming research.
Red-team this investment thesis: [THESIS]. Review these sources and facts: [SOURCE MATERIALS]. My assumptions are [ASSUMPTIONS], and the decision horizon is [HORIZON]. Produce the five strongest alternative explanations, missing evidence, historical or peer comparisons worth checking, scenario assumptions most likely to fail, and explicit falsification signals. Do not invent evidence, citations, or prices. Ask up to 3 questions only if required inputs are missing. Check that every critique links to supplied evidence or is labeled as a research question, not a fact.
Tips
- Use source dates for every financial figure because revenue, share count, guidance, and holdings can change between reporting periods.
- Keep an assumptions ledger separate from extracted facts; a clean model can still be misleading if growth, margins, discount rate, or terminal assumptions are implicit.
- Read risk factors, related-party disclosures, debt terms, customer concentration, and share-based compensation instead of relying only on headline revenue growth.
- A scenario model is most useful when the downside case changes a real driver, such as churn, pricing, margins, refinancing, or dilution, rather than merely lowering the final valuation.
FAQ
Can AI tell me what stock to buy?
It should not make personalized buy or sell decisions. It can help organize verifiable research, test assumptions, and identify questions for a licensed financial professional.
What sources should I provide?
For public companies, start with annual and quarterly filings, earnings materials, audited financial statements, and official disclosures. For funds, use the prospectus, fact sheet, holdings file, and regulatory filings.
How should I use a valuation output?
Treat it as a sensitivity exercise, not a precise target. Check the input periods, units, formulas, and assumptions, then compare the result with the risks and your own portfolio context.