varunk130 OKR Cascade Planner
Turns strategy into a coherent cascade of Objectives and Key Results - without becoming a task list.
What this skill is
A workflow that translates a company strategy into Objectives and Key Results (OKRs) at three levels (company → function → team), enforces vertical and horizontal alignment, distinguishes leading from lagging Key Results (KRs), sets honest scoring conventions, and installs the quarterly rituals that make OKRs work. Built to avoid the two most common failures: KRs that are tasks, and cascades that become wishlists.
What it solves
- OKRs that are restated task lists with no measurable outcome
- Horizontal misalignment (teams pulling in conflicting directions)
- Sandbagging - KRs set low so 100% is easy
- Quarterly reviews that turn into status updates
- Functions copying each other's OKRs to look aligned without integration
When to invoke
- First-time OKR rollout
- Fixing a broken OKR program after Q1-Q2 disappointment
- Annual planning to quarterly OKR translation
- Onboarding a new team or function to OKR mechanics
- Pre-quarter-end review to score and reset
Phase 1: Anchor on company strategy
Before writing any OKR, get clear on:
- The company's 3-year strategic direction (1 paragraph)
- The 3-5 strategic priorities for the year
- The constraints (capital, talent, time)
If strategy is fuzzy, OKRs will be too. Push back upward before starting.
Phase 2: Write company-level Objectives
Each Objective must be:
- Qualitative - describes the change, not the metric
- Inspirational - worth doing
- Time-bound - quarter or year
- Action-oriented - starts with a verb
Bar: 3-5 company Objectives. More than that, there are no priorities.
Good: Become the default choice for mid-market customers in our category. Bad: Improve revenue. (vague) / Ship feature X. (output, not outcome)
Phase 3: Write Key Results (the test)
For each Objective, write 3-5 Key Results.
KR test (every KR must pass):
- Measurable - single number you'd bet on
- Time-bound - by end of quarter or year
- Outcome, not output - moves a real-world metric, not "did we ship X"
- Set the bar credibly - at approximately 70% confidence; 100% should mean stretch
- Owned by a single person - multiple owners = no owner
Bad: Launch new onboarding. (output) Better: Reduce time-to-first-value from 14 days to 5 days.
Mix leading and lagging KRs:
- Lagging - confirms direction (Net Revenue Retention (NRR), Annual Recurring Revenue (ARR) growth) - slow signal
- Leading - predicts (activation rate, demo conversion) - fast signal
A KR set with only lagging measures arrives too late to course-correct.
Phase 4: Cascade to function and team
Each function (and each team within) writes 2-4 Objectives that support at least one company Objective.
Cascade rules:
- Function and team OKRs support company OKRs - they are not copies
- Each function or team identifies its unique contribution to the parent Objective
- A function may have its own functional Objectives (e.g., engineering quality) that don't ladder up - that's fine; cap at 1 per function
Build the alignment map:
| Company OKR | Function OKR | Team OKR | Owner |
|---|---|---|---|
| Default choice in mid-market | Sales: win-rate up in mid-market | Sales Development Representative (SDR) team: qualified pipeline up 2x | Name |
| Product: mid-market feature parity by Q3 | Platform team: Single Sign-On (SSO) plus audit log Q2 | Name |
Phase 5: Horizontal alignment check
The cascade reveals dependencies. Before committing:
- Does any team's KR depend on another team's KR being hit?
- If yes, the dependency must be explicit and the upstream team must accept the commitment
- Resolve conflicts now, not at quarter end
Map the dependency graph. Single biggest miss in OKR programs: KRs hit individually, outcomes still missed because dependencies weren't aligned.
Phase 6: Set scoring conventions
Decide upfront - and write it down - what scores mean:
| Score | Meaning |
|---|---|
| 1.0 | Stretch goal fully delivered (rare; suggests sandbagging if frequent) |
| 0.7 | Strong execution at the originally set ambition |
| 0.4 | Genuine progress but the goal wasn't reached |
| 0.0 | No meaningful progress |
Target average: 0.6-0.7 across the organization. Average too low → over-set. Average too high → under-set.
Compensation linkage: keep OKRs out of compensation formulas. Tying directly creates sandbagging.
Phase 7: Quarterly rituals
Four rituals make OKRs work:
| Cadence | Ritual | Duration | Output |
|---|---|---|---|
| Weekly | Team check-in on KR progress | 15 minutes | Red-Yellow-Green (RYG) status, blockers |
| Monthly | Cross-function alignment review | 60 minutes | Risks, re-prioritization, dependency unblocking |
| Quarter-end | Scoring plus retrospective | 90 minutes | Honest scores, what worked and what didn't |
| Pre-quarter | Drafting plus alignment | 2 weeks | New OKRs committed |
No drafting committee. No 6-week planning process. Two weeks maximum.
Phase 8: Anti-patterns to watch
- KRs as tasks - "Launch X" → fail
- Too many OKRs - more than 3 Objectives per team kills focus
- Set-and-forget - no weekly check-in → drift
- Re-baseline mid-quarter - undermines the discipline
- Cascade as copy - every team has the same KRs → no division of labor
Output
- Company Objectives (3-5) with KRs (3-5 each)
- Function and team OKRs with alignment map
- Dependency graph with accepted commitments
- Scoring rubric and target average
- Weekly, monthly, and quarterly ritual calendar
- Pre-mortem: where could the cascade fail?
- Re-planning trigger: what changes would force a reset?
Operating rules
Always
- Anchor every Objective in a stated strategic priority
- Make every KR a measurable outcome, not an output
- Mix leading and lagging KRs
- Make dependencies explicit and accepted
- Score honestly (0.6-0.7 average is healthy)
Never
- Use OKRs as a task list
- Tie scoring directly to compensation
- Run more than 3 Objectives per team
- Cascade by copying parent KRs
- Re-baseline mid-quarter to hit scores