thinkhard.SI

Prompt repository Managers & leads

Review your team's draft OKRs before the quarter starts

Your team has drafted objectives and key results for next quarter, and you want them checked for vague goals, activity-style key results and overload.

Registry no.
MGT-053
Collection
Managers & leads
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
6

Managers & leadsAny chat

The prompt

Score 10 of 10, Excellent

Use it when Your team has drafted objectives and key results for next quarter, and you want them checked for vague goals, activity-style key results and overload.

  • {{team_name}}
  • {{deadline}}
  • {{priorities}}
  • {{draft_okrs}}
  • {{capacity}}
  • {{baselines}}
Full prompt and why it works
Prompt
Review the draft objectives and key results (OKRs) below for {{team_name}} before we commit to them on {{deadline}}. I will bring your review to a working session with the team, so suggest a fix for each problem you find, and leave the final choices to us.

Company or department priorities for the quarter:
"""
{{priorities}}
"""

Our draft OKRs:
"""
{{draft_okrs}}
"""

Team capacity next quarter, and work already committed outside these OKRs: {{capacity}}
Baselines we have for the numbers in the key results, with where each comes from: {{baselines}}

Check each objective and key result against these tests:
- Outcome, not activity: a key result says what will change, not what we will do. For example, "Ship the new onboarding flow" is an activity; "Raise the share of new accounts active in week two from [baseline] to [target]" is an outcome.
- Measurable at the end of the quarter: a baseline and a target, each with a named source.
- Within our control or strong influence, not mainly dependent on another team.
- Linked to a priority above, named in brackets.
- Ambition: say if a target looks like a sure thing or out of reach, given its baseline, and why.

Then:
1. A table with the columns Item, Problem, Suggested rewrite, Question for the team. Include only items with a problem.
2. Overload check: whether the set fits our capacity once committed work is counted, and which key result to drop first if it does not.
3. Dependencies on other teams that need agreeing before the quarter starts.

Keep the review under 600 words, in a direct, constructive tone. Do not invent baselines or targets; where a key result has no baseline, suggest making "measure the baseline" its first step and mark the number [BASELINE NEEDED]. Do not add new objectives unless a priority above has none, and then propose at most one. A good review leaves the team with a short set of OKRs they could score honestly at the end of the quarter.

Variables

{{team_name}}
Your team, for example "the customer onboarding team".
{{deadline}}
When the OKRs are due to be final, for example "Friday, March 28".
{{priorities}}
The company or department priorities for the quarter, as written.
{{draft_okrs}}
The team's draft objectives and key results, as written.
{{capacity}}
People and time available next quarter, and work already committed, for example "4 people, one on leave in May, plus the platform migration".
{{baselines}}
Current numbers for the key results and where each comes from, or "none yet".

Why it works

  • What good looks like. Five named tests, including outcome rather than activity and a baseline with a source, tell the AI exactly what makes a key result weak.
  • An example. A paired example of an activity and an outcome shows the rewrite wanted, without supplying the team's real numbers.
  • If unsure. Missing baselines are marked [BASELINE NEEDED] and turned into a first step, never filled with an invented figure.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
  4. When the answer comes back, check it before you use it. The Fact Check List lists every figure, quote and source in the answer worth checking, riskiest first.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/okr-draft-review

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "okr-draft-review"}, or just ask for it by name.