I need a research-and-analysis brief on [RESEARCH QUESTION] for [AUDIENCE AND DECISION]. Use the scope, time period, geography, and definitions in [RESEARCH SCOPE]. Start with the sources and notes I provide in [SOURCE MATERIAL]. If web access is available, supplement them only with credible, current sources; prefer primary documents, official statistics, filings, peer-reviewed research, and direct company disclosures over commentary. Method: 1. Restate the question as 3-5 answerable subquestions. Identify the key terms, comparison group, and evidence needed for each. 2. Separate facts, estimates, interpretations, and recommendations. Cite every externally verifiable factual claim inline with a source name, publication date, and link where available. 3. Assess source quality, recency, sample limitations, conflicts of interest, and whether sources agree. Do not treat a correlation as causation.
4. Analyze implications against [EVALUATION CRITERIA], including counterarguments and plausible alternative explanations.
Produce a 900-1,200 word brief with: Executive answer (5 bullets); Scope and method; Evidence table with claim, source, date, strength, and caveat; Analysis; Options or implications; Open questions; and Recommended next research steps. Include a short source list with links.
For any financial, investment, legal, medical, or safety implication, present information and risks, not personalized professional advice. Do not invent data, quotes, citations, access to sources, or certainty.
Before finalizing, run two checks: flag every conclusion that rests on weak, old, incomplete, or conflicting evidence; and verify that each recommendation follows from cited evidence rather than assumption. Ask up to 3 clarifying questions only if a required input is missing.Fill in
| Placeholder | What to enter | Example |
|---|---|---|
| [RESEARCH QUESTION] | Enter the precise question you need answered. | Should a regional grocery chain add same-day delivery in its Ohio stores? |
| [AUDIENCE AND DECISION] | State who will read the brief and what decision it will inform. | COO deciding whether to fund a 12-month delivery pilot. |
| [RESEARCH SCOPE] | Define the relevant time period, geography, market, population, and exclusions. | Ohio metropolitan stores; 2024-2026 data; exclude pharmacy delivery. |
| [SOURCE MATERIAL] | Paste links, documents, data, interview notes, or prior research to assess. | 2025 customer survey, store labor report, Instacart proposal, competitor pricing screenshots. |
| [EVALUATION CRITERIA] | List the criteria that should determine the conclusion or recommendation. | Incremental profit, customer retention, labor impact, operational risk, and time to launch. |
How to use
- Paste the prompt and replace every bracketed input with your question, decision context, and source material.
- Attach spreadsheets or reports when possible, and tell the model which source should control if figures conflict.
- Check the evidence table before relying on the executive answer; open the cited links for high-stakes claims.
- Follow up with: “Turn the open questions into a research plan with an owner, source, method, and deadline for each.”
Variations
Competitor brief
Use this to compare named competitors before a market decision.
Analyze [COMPETITORS] in [MARKET] for [DECISION MAKER]. Use [SOURCES OR LINKS] and distinguish verified facts from inference. Compare target customer, offering, pricing, distribution, positioning, growth signals, and material risks in a table. Then write a 600-word assessment of where [OUR COMPANY] is differentiated, vulnerable, or missing evidence. Cite factual claims inline and include a source-quality note. Do not invent pricing, traction, or strategy details. Ask up to 3 questions only if essential inputs are missing.
Stock research memo
Use this for a factual company-research memo, not investment advice.
Prepare a neutral research memo on [PUBLIC COMPANY] for [TIME HORIZON]. Analyze [EARNINGS MATERIALS], [FILINGS OR LINKS], and [COMPARISON COMPANIES]. Cover business model, revenue drivers, unit economics where disclosed, valuation inputs, balance-sheet and dilution risks, management guidance, catalysts, and bear cases. Cite each factual claim with a dated source. Produce a one-page memo plus a table of metrics and unknowns. Do not give personalized investment advice or invent financial figures. Ask up to 3 questions only when required inputs are missing.
Data findings report
Use this after you have a dataset and need analysis suitable for stakeholders.
Analyze [DATASET DESCRIPTION] to answer [BUSINESS QUESTION] for [AUDIENCE]. Use [DATA DEFINITIONS AND TIMEFRAME] and the calculations or constraints in [ANALYSIS RULES]. First identify missing values, duplicate records, denominator issues, outliers, selection bias, and metrics that cannot support causal claims. Then produce a findings report with methodology, 5 key findings, a results table, segment comparisons, limitations, and decisions the data can and cannot support. Show formulas or calculation logic in plain language. Do not fabricate rows or results. Ask up to 3 questions only if required data details are absent.
Tips
- Give the model the decision being made, not just the topic; it changes which evidence and tradeoffs matter.
- Treat publication date as part of the evidence, especially for market shares, regulations, prices, and company performance.
- Ask for an evidence table before accepting a narrative conclusion; it exposes where a claim is based on one weak source.
- For research with stakes, verify primary sources yourself and preserve the links and dates used in the final brief.
FAQ
Can AI research a topic from the web on its own?
It can help find and synthesize sources when browsing is available, but you should still verify important citations and dates. Give it trusted starting sources when accuracy matters.
How do I stop it from making up citations?
Require inline citations, a source table, and an explicit uncertainty section. Ask it to say “not found” rather than fill gaps.
Can I use this for investment research?
Yes, for structured factual research and risk analysis. It should not replace a licensed professional or be treated as personalized investment advice.