Feedback — Radical Candor
Deliver feedback that the recipient can hear AND act on. Specific, kind, future-focused.
Prompt
**Role:** Engineering manager who has read Radical Candor 4 times and practiced it for 8 years. You know that "kind" without "candid" is ruinous empathy, and that "candid" without "caring" is obnoxious aggression. **Context:** Person: [name + role + how long you've worked together]. The specific behavior: [what they did or didn't do]. Impact: [what happened because of it — team, customer, themselves]. Pattern: [is this a one-off or a pattern?]. Their stated growth area: [if known — what they're working on]. **Task:** Write the feedback as a 5-min 1:1 script. 1. The opening (1 sentence): name what you want to talk about. NOT "How are things?" but "I want to talk about the team review meeting yesterday." 2. The specific (2-3 sentences): the observable behavior. Quote what was said or describe what was done. No interpretation yet. 3. The impact (1-2 sentences): what happened because of it — for the team, the customer, them. Tie it back to something they care about. 4. The question (1 sentence): "Help me understand what was going on for you" or "What were you optimizing for?" — open-ended, gives them space to respond. 5. The path forward (1-2 sentences): the small change you're asking for + when you'll check back in. 6. The care (1 sentence): why you're investing this conversation in them. "I'm telling you this because I think you can be exceptional at X." **Constraints:** - Lead with the specific, not the general - "I noticed" / "I observed" — not "you always" / "you never" - Care is at the END (recency), not the beginning (where it sounds like a cushion) - One behavior per feedback — don't pile on - The path forward is a small step, not a transformation **Output format:** 6-sentence-ish script · plus 1-paragraph "if they get defensive" recovery line.
How to use this prompt
- 1
Copy the prompt above and paste it into ChatGPT, Claude, or Gemini — or open it in the visual Studio to edit each part on a canvas and run it with your own key.
- 2
Replace any bracketed placeholders with your specifics. The more concrete your context and constraints, the sharper the result — see the 5-part prompt structure.
- 3
Run it, then refine. Ask the model to critique and improve its own answer with self-critique prompting.
Techniques in this prompt
Assigns the model an expert persona so it adopts the right vocabulary, depth, and standards for the task.
Learn this techniqueSets the rules and boundaries — tone, length, what to avoid — that keep the output on-target.
Learn this techniqueSpecifies the exact shape of the result — sections, a table, JSON, a word count — so the output is predictable and ready to use.
Learn this techniqueRecommended models
Build on this prompt
Open it in the visual Studio to wire it into a full workflow with your own API key — or learn the craft behind prompts like this.