Making sense of staff surveys
Listen to this lesson
Nigerian voice narration
Fictional training case. 150 staff submitted open-ended welfare survey comments and management wants the top three themes by next week's meeting. Manually reading 150 free-text responses is slow and inconsistent between readers, so AI is used to group comments into themes while keeping every respondent anonymous. Work through the example, identify the evidence needed and practise the method with cleared or synthetic inputs. Any additional figures in the explanation are separate practice examples, not claimed rows in a downloadable file.
Worked example
Remove names and any identifying phrases ('the officer in Procurement who...') from the raw comments before pasting them. Ask AI to cluster the comments into themes, report how many comments fall under each theme, and include one representative quote per theme so management can see real voice, not just a label. Cross-check the counts against the raw file — a theme claimed as the 'top concern' should actually have the highest comment count. Then ask for 3 low-cost, practical recommendations per theme suited to a public-sector budget, and verify none assumes a budget or authority the department does not have.
The method
Theme-then-count-then-quote: ask for a structured three-part output (theme name, count, quote) instead of a free summary, so you can audit the AI's grouping against the raw data rather than trusting an unverifiable paragraph.
Worked practice and common mistakes
A comment can mention workload and training at once. Define whether multiple coding is permitted; if so, theme counts can exceed response counts and are not separate people. Keep minority and positive comments as well as complaints. Verify each quote against its source ID rather than accepting an invented representative quote. Counts, definitions and denominator must remain visible together.
Review before using the result
Recount two of the claimed themes manually from the raw comments. If the numbers do not match, ask AI to re-cluster with tighter theme definitions before the recommendations go to management.
Your workplace practice
Use the recommended free assistant with this lesson's fictional example. Ask for the output described in the worked example. Keep a copy of the input and draft. Compare the result with the verification steps above, record any correction, and retain placeholders where evidence is missing. The expected result is a checked draft for human review, not an automatically approved decision. Course examples and practice responses are fictional; amounts, dates, rules and thresholds are not official guidance or supplier quotations.
Why this tool: Can suggest survey themes from fictional samples, with counts you verify manually.
Hit a free limit? Switch to ChatGPT or Gemini and continue — the prompt works in any assistant.