Performance review prompt builder
Performance reviews go wrong when they are vague, biased toward recent events, or inconsistent between managers. This tool builds an AI prompt that forces the opposite: specific, evidence-based feedback tied to your competency framework and rating scale, ending in measurable goals for the next period.
How it works
You choose whether you are writing a manager review or a self review, pick your rating scale (1-5, 1-3, labelled, or narrative-only), and list the competencies your organisation assesses against. The builder assembles a prompt that, for each competency, requires:
- A rating on your chosen scale.
- Two to three sentences of behaviour-based feedback citing concrete examples.
- One clear, actionable development suggestion.
The prompt explicitly tells the model to avoid personality judgements, to cover the entire review period rather than only recent weeks, and to finish with a short set of measurable goals. The result is a structure that produces defensible, useful reviews instead of generic praise.
Tips and examples
The single most important step is to paste your own notes after the prompt. The tool produces the scaffolding; the evidence has to come from you — a few bullet points per competency (“shipped the billing migration two weeks early”, “unblocked two teammates during the outage”) is enough for the model to write grounded feedback. Keep a running log through the period so you are not reconstructing it from memory at review time, which is where recency bias and fabrication creep in. Always read and edit the output before it goes into a formal record.