Library › Research AI › Managers and team leads

Check your team deck's claims against the engagement survey

Practice: Catching mistakes and made-up facts. About 5 minutes.

About this exercise

You drafted 5 slide headlines about your team's engagement survey and want to check them before the all-hands. You want exactly 5 numbered lines, one per headline, each labeled Supported, Overstated, or Not in survey, with the survey figure after any Overstated label. Use only the survey results, plain text, and nothing else.

Common mistakes when you ask AI for this

  • Asking if the slides are good. The AI gives presentation tips and rewrites headlines instead of checking them. Try: Say: "Label each headline against the survey only."
  • No fixed labels. The AI hedges with 'mostly accurate' and 'could be clearer', which you can't act on. Try: Give them: "Supported, Overstated, or Not in survey."
  • Missing a flat trend. "Climbing" sounds fine unless you compare to the spring number, which the AI may not do. Try: Add: "Compare to spring numbers when a headline claims a change."
  • Not saying how many. The AI decides the length itself and usually returns more lines, one per headline than you can use. Try: Be exact: "exactly 5 numbered lines, one per headline."
  • Getting markdown back. AI defaults to bold labels and headings, which show up as stray symbols in your speaker notes. Try: Ask for "plain text, no bold, no headings."

What a strong prompt includes

  • Gives exact labels
  • Names the source
  • Checks trend claims
  • Says exactly how many
  • Asks for plain text

Practice this now ↑

Related exercises