Library › Leadership AI › Managers and team leads

Draft review evidence in Situation, Behavior, Impact form

  • For Managers and team leads
  • You practice Keeping the AI to your sources
  • Time about 5 minutes
  • Level Intermediate

About this exercise

You're drafting the evidence section of Leo's annual review, which HR reviews before anything is shared. Write exactly 3 numbered examples, each with the labels Situation:, Behavior:, Impact:, using only dated events from your notes. Don't suggest a rating and don't use personality words. End with the line "Draft for HR review, not final." Under 200 words.

Common mistakes when you ask AI for this

  • Copying your opinions. The AI repeats 'brilliant but arrogant' because it's in your notes. Try: Add: "Describe behavior only. No personality words like arrogant or abrasive."
  • Letting the AI pick a rating. The AI concludes with 'Exceeds expectations' to sound complete. Try: Add: "Don't suggest a rating; that's my call after calibration."
  • Vague examples. The AI writes 'Leo is a strong collaborator' without a date or event. Try: Ask: "Each example has Situation:, Behavior:, Impact: and a date from the notes."
  • No review marker. The draft looks final and could be pasted straight into the system. Try: Add: "End with: Draft for HR review, not final."
  • Too many examples. The AI turns every note into an example. Try: Say: "Exactly 3 numbered examples."
  • Trusting the AI to count words. AI is bad at counting. Ask for a total word limit and it often lands over it. Try: Cap each part instead: "each line under 15 words."

What a strong prompt includes

  • Names the SBI labels
  • Fixes the count
  • Bans personality words
  • Keeps the rating out
  • Marks it for HR

Practice this now ↑

Related exercises