writing-okrslisted

# Writing OKRs ## Route by task | User intent | Workflow | |---|---| | Draft new OKRs from provided context | [Write workflow](#write-workflow) | | Improve existing draft OKRs | [Refine workflow](#refine-workflow) | | Score or audit existing OKRs | [Evaluate workflow](#evaluate-workflow) | --- ## Write workflow 1. **Structure the output using the template.** Load [references/03-templates.md](references/03-templates.md) and fill every field. Never omit `Parent OKR`, `Type`, or `Why this matters`. 2. **Draft Objectives.** Each must be qualitative (no numbers), directional, and pass the "Who Cares?" test: *if we 1.0'd this, would a customer, investor, or employee notice?* If not, reword or cut. 3. **Draft Key Results using the 4-part formula:** ``` [metric name] from [baseline] to [target] by [date] ``` If a baseline is unknown, use an explicit ⚠️ placeholder — never silently omit it. Format: `⚠️ [replace with actual — e.g., 88%]`. A KR with an unfilled baseline cannot be scored at mid-cycle. 4. **Apply the two litmus tests to every KR:** - **Output vs. outcome:** Can the KR be checked off without a metric moving? If yes, it's a task — rewrite as an outcome. - **Necessary but not sufficient:** If the team scored 1.0 on every KR, is it possible the Objective still wasn't achieved? If yes, the KR set is insufficient — add or rewrite KRs until their collective success guarantees the Objective. 5. **Tag every OKR as committed or aspirational** (see [Commit