Intermediate 20-30 minutes
Turn Survey Comments Into Auditable Themes and Actions
Create a staged analysis that separates anonymization, coding, theme validation, and recommendations.
Recommended tools: ChatGPT, Claude, Google Gemini
Live Preview
ROLE
Act as a qualitative research workflow designer.
OBJECTIVE
Create a prompt chain for analyzing [SURVEY RESPONSES] to inform [DECISION GOAL].
CONTEXT
Survey questions: [QUESTION TEXT]. Audience context: [AUDIENCE CONTEXT]. Privacy requirements: [PRIVACY RULES].
INSTRUCTIONS
Design stages for removing unnecessary identifiers, creating a provisional codebook, coding responses, checking coding consistency, and synthesizing themes. Write the actual prompt for each stage with explicit input and output contracts. Preserve anonymous response identifiers so themes can be traced to evidence. Allow multiple codes, ambiguous classifications, and contradictory views. Separate respondent counts from mention counts and use deterministic counting where possible. Do not equate frequency with importance or generalize beyond the sample. Keep recommendations distinct from participant statements and include an evidence check before action planning.
OUTPUT FORMAT
Provide stage prompts, a codebook template, a coded-response schema, a theme evidence table, and an action brief format. Demonstrate the chain on a small fictional example.
QUALITY CHECK
Flag assumptions and avoid fabricated statistics. Explain how quotations will be checked against originals and how identifying details will be excluded.
Open in:We copy the prompt first, then open the tool in a new tab. Results vary by tool.