OKR Refactor — From Wishful to Falsifiable

**Role:** Operations lead who has run 8 quarters of OKR cycles and seen which OKRs drive execution vs which become wallpaper.

**Context:** Current draft OKRs: [paste the team's wishful version]. Quarter: [Q-N timeframe]. Team size + capacity: [N people, M weeks of focus time]. The strategic constraint: [the one thing that, if hit, makes the quarter a win].

**Task:** Refactor each Objective + Key Results so they're outcomes (not outputs), measurable, and bet-able.

1. For each Objective: ensure it's an outcome (a state of the world) not an output (a thing we ship). NOT "Launch v2" but "v2 drives 30% week-2 retention lift."
2. For each Key Result: it must be NUMERIC + TIME-BOUND + ATTRIBUTABLE. NOT "Increase NPS" but "NPS from 32 to 45 by EOQ."
3. Stress-test: for each KR, what's the test that proves it's done? If the test is "we feel good about it," the KR fails.
4. Identify the ONE KR that, if hit, makes the quarter a win regardless of the others. That's the focus.
5. Cut anything that isn't a bet. If we don't really care if it ships, it doesn't belong in OKRs.

**Constraints:**
- Each KR has a number, a baseline, and a target
- Each KR has an owner (one person)
- No "increase / improve / optimize" without a number
- No more than 3-4 KRs per Objective

**Output format:** Refactored OKRs table · before/after for each · plus 1-paragraph "which one is the focus and why" callout.