Home › Compare › Research report
Research report: same prompt, different AI
Task: A sourced report on a question you need to understand. Below is the same job written the way each tool works best. All prompts are RuleStack originals (MIT). We describe how prompting differs, not which tool is "best": we haven't run benchmarks, and results change with every model update.
Note: Review and test before use. Provided as is, without warranty. See Terms.
Perplexity
Perplexity cites inline by default; ask for a numbers table to make checking easy.
Research [question] for [audience]. Use Research mode. Structure: summary, key findings with inline citations, conflicting views, a table of key numbers with source and date, open questions. Prefer primary sources.Gemini
Gemini Deep Research shows a plan you can edit before it runs.
Run Deep Research on: [question]. Audience: [audience]. Before starting, show me your research plan so I can adjust it. Final report: executive summary, findings with sources, limitations, recommendations.ChatGPT
ChatGPT deep research often asks clarifying questions first; answer them for a sharper report.
Use deep research to answer: [question]. Scope: [regions, time period]. Ask me up to 3 clarifying questions first. Report with sections, citations, a confidence note per finding, and a list of sources.Claude
Asking Claude to separate facts from interpretation makes reports easier to trust.
Research [question] using web search. Work in steps: list sub-questions, search each, then synthesize. Cite sources inline, separate facts from your interpretation, and end with what evidence would change the conclusion.How they differ
- All research modes can miss or misread sources. Open the top sources before you rely on a claim.
- Plans and clarifying questions improve results; say what you will use the report for.
- Ask for dates on every number, since research tools mix old and new data.
Replace everything in [square brackets] with your details. More: Research prompts · all comparisons