Home / Data & Analysis / Research & Synthesis

Reliable Research Workflow in 7 Phases, from Plan to Final Audit

Takes a topic through a full research process: a plan of key questions, web research with cited sources, sorting facts from opinion, finding gaps, weighing sources that disagree, extracting 5 insights and a critical audit before you rely on the results.

researchsourcesfact checkinginsightsverification
  1. Phase 1 — Research plan
    I want to research [topic]. Help me plan the work before I start.

1. Split the topic into the key questions I must answer.
2. For every question, explain:
   - why it is important
   - which kinds of sources I should look for
   - what evidence would make the answer trustworthy
3. Finally, suggest the most effective order in which to tackle the questions.

Present the questions as a numbered list.
  2. Phase 2 — Research with quality sources
    Use web search to research [topic].

Give priority to: primary sources, official documents, original studies, academic papers, government statistics and reputable outlets.

Back every major claim with a citation.
  3. Phase 3 — Facts versus opinions
    Take the findings on [topic] (from our research above, or pasted here: [findings]) and sort them into four groups:
1. Confirmed facts
2. Strong evidence
3. Reasonable interpretations
4. Uncertain or disputed claims
  4. Phase 4 — Gaps in the research
    Here is my research: [paste research or refer to the research above].

Find what it lacks:
- Which important questions are still unanswered?
- Which assumptions am I making?
- Which viewpoints, stakeholders, data sources or time periods have I left out?

Return the biggest gaps as a list ranked by priority, and for each one explain how I could investigate it.
  5. Phase 5 — Sources that disagree
    My sources do not agree on [claim].

Compare them on:
- the evidence they use
- their methodology
- when they were published
- possible biases
- overall source quality

Then explain where they agree, where they diverge, and which conclusions have stronger support.
  6. Phase 6 — Actionable insights
    Working from this research ([paste research or refer to the research above]), give me 5 insights I can act on.

For each insight include:
- the evidence behind it
- why it matters
- an action that could follow from it
- its biggest uncertainty or risk
  7. Phase 7 — Final quality audit
    You are a critical reviewer of research. Before I use the research below, audit it: [paste research or refer to the research above].

Look for:
- claims without support
- weak sources
- outdated information
- missing counterarguments
- gaps in logic
- contradictions
- conclusions stated with too much confidence

Flag every point that still needs to be verified.
More in Research & Synthesis
15 Expert Questions and Answers About a PDFNewResearch & SynthesisPlays the top expert on the subject and writes the 15 key questions a sharp reader should ask after the PDF, each answered in detail with evidence, deeper implications and overlooked insights, ending with the 3 that matter most.
Role: you are the foremost expert in the world on the subject of the PDF attached to this conversation. If nothing is attached, ask me to upload it first.

Task: write the 15 most important questions that an intelligent reader should ask after reading this document.

For every question:
- give a clear and detailed answer
- refer to the evidence in the document that supports it
- explain what it implies at a deeper level
- point out any insight that is surprising or easy to miss

End with:
- the 3 questions that matter most
- the single most important answer
- the biggest idea an average reader would probably miss
Open page
Academic-Style Research on Any Topic in 7 Phases7 phasesNewResearch & SynthesisTakes one research topic through seven phases: setting up a rigorous research mindset, defining scope, mapping the knowledge landscape, exploring core concepts, critical evaluation, advanced research questions and a final academic-style report.
  1. Phase 1 — Research mode
    Role: you are a senior research professor. You trained at Stanford University and you are respected for the rigour of your academic research.

From now on, each time I give you a research topic, tackle it methodically: identify its key themes, its core concepts and the important questions the research must answer.

My research topic: [topic]
  2. Phase 2 — Research scope
    Role: you are a professional academic researcher.

Topic: [topic]

Task: help me pin down the precise scope of my research on this topic. Present it under three headings:
1. Subtopics
2. Boundaries (what is in and what is out)
3. Key research questions that have to be answered
  3. Phase 3 — Knowledge landscape
    Role: you are a research analyst.

Topic: [topic]

Task: map the whole knowledge landscape of this topic, so I can see how the field is organized. Identify:
- the major themes
- the subfields
- the foundational ideas
- the trends that are emerging
  4. Phase 4 — Core concepts in depth
    Topic: [topic]

Task: explain in depth the core concepts of this topic. Define each concept clearly and show how it connects logically to the others, so I build an understanding that holds together.
  5. Phase 5 — Critical evaluation
    Topic: [topic]

Task: evaluate this topic critically. Point out:
- the assumptions it rests on
- its logical weak spots
- the questions that remain unanswered
- the areas that need more investigation
  6. Phase 6 — Advanced research questions
    Topic: [topic]

Task: write research-level questions about this topic that push thinking further and lead the exploration well past a basic understanding.
  7. Phase 7 — Research report
    Topic: [topic]

Task: write a complete report on this topic in research style, so it reads like professional academic work. Use these sections:
1. Introduction
2. Key findings
3. Analysis
4. Conclusion
Open page
Bull vs Bear vs Fact-Checker: Three Sub-Agents Debate a TopicNewResearch & SynthesisLaunches three sub-agents on a market trend or topic: one builds the strongest bull case, one the strongest bear case, one checks their sources. They debate, and you get only the objective conclusion that survives.
Topic to analyze: [MARKET TREND/TOPIC]

Set up three sub-agents:
- Agent 1: builds the strongest possible bull case.
- Agent 2: builds the strongest possible bear case.
- Agent 3: fact-checks the sources both of them use.

Make the agents debate each other. Show me only the final conclusion, objective and the one that still stands after the argument.
Open page
Decision-Ready Executive Brief from a DocumentNewResearch & SynthesisReviews a document as a Chief Strategy Officer and writes an executive briefing that covers what is happening, why it matters, evidence, risks, opportunities, implications and actions, closing with top 3 priorities, the most urgent decision and the next step.
Role: you are a Chief Strategy Officer reviewing the document attached to this conversation. If nothing is attached, ask me to upload it first.

Task: write an executive briefing that is ready for decision-making, organized under these questions:
1. What is happening?
2. Why does it matter?
3. Which evidence supports the conclusion?
4. What are the main risks?
5. Which opportunities should be seized?
6. What are the strategic implications?
7. Which actions should be taken?

End with:
- the 3 top priorities
- the decision that is most urgent
- the opportunity with the biggest impact
- the next step you recommend
Open page
Evidence-Based Research Summary: Facts vs OpinionsNewResearch & SynthesisResearches a topic, splits confirmed facts from opinions, weighs the strongest sources against each other and returns a short summary grounded in evidence.
Topic: [topic]

Task:
1. Research the topic.
2. Keep confirmed facts and opinions clearly apart.
3. Compare the most reliable sources with one another.
4. Write a concise summary based on the evidence, and show which source supports each confirmed fact.
Open page