Product OKRs: How to Write OKRs for Product Teams (With Examples)
Learn how to write effective OKRs for product teams. Includes real examples, common mistakes, and templates for product managers at every level.
What OKRs Are (And Aren't)
OKRs—Objectives and Key Results—are a goal-setting framework popularized by Intel and Google:
| Component | Description |
|---|---|
| Objectives | Qualitative descriptions of what you want to achieve |
| Key Results | Quantitative measures of success |
What OKRs Are NOT
- A to-do list
- Project plans
- Descriptions of what you'll build
"Launch feature X" is a task. "Improve new user activation to 40%" is an OKR.
The purpose is alignment and ambition. OKRs ensure everyone understands what success looks like and pushes toward stretch goals rather than safe targets.
Anatomy of a Good Objective
Objectives are qualitative and inspirational. They should be memorable and motivating:
- "Delight new users from day one"
- "Become the trusted platform for enterprise teams"
- "Expand into international markets successfully"
Good Objectives Are:
- Clear — everyone understands them
- Directional — they guide decisions
- Ambitious — they push beyond the comfortable
Good Objectives Are NOT:
- Vague ("do a good job")
- Metric-focused (save numbers for KRs)
- Boring (they should energize people)
Limit to 3-5 objectives per quarter. More than that and you've lost focus. If everything is a priority, nothing is.
Anatomy of Good Key Results
Key Results are quantitative and measurable. Each Objective should have 2-4 Key Results:
- "Increase day-7 retention from 20% to 35%"
- "Reduce time-to-first-value from 10 minutes to 3 minutes"
Good KRs Are:
| Quality | Description |
|---|---|
| Specific | Exact numbers |
| Measurable | You can track them |
| Outcome-focused | Results, not activities |
| Time-bound | Achievable within the OKR period |
Avoid Activity-Based KRs
❌ "Ship 5 features" or "Run 10 experiments"
These are outputs, not outcomes. What happens if you ship 5 features but none improve anything? The KR is met but the Objective isn't.
Stretch Goals vs. Commitments
OKRs should be stretch goals—achievable but not easy. Google aims for 70% completion as success.
If you hit 100% of your OKRs, they weren't ambitious enough.
The Accountability Tension
If 70% is success, how do you hold people accountable?
Judge effort and progress, not just achievement. A team that pushed hard and reached 70% did better than a team that sandbagged and hit 100%.
Committed vs. Aspirational
Some organizations use a mix:
| Type | Meaning |
|---|---|
| Committed OKRs | You must hit these (floor targets) |
| Aspirational OKRs | Stretch goals you might not reach (ceiling targets) |
Product Team OKR Examples
New User Success
Objective: New users succeed from day one
| Key Result | Target |
|---|---|
| Day-7 retention | 15% → 25% |
| Time to first value | 8 min → 3 min |
| Onboarding completion rate | 40% → 60% |
Enterprise Growth
Objective: Enterprise customers choose us over competitors
| Key Result | Target |
|---|---|
| Win rate on enterprise deals | 20% → 35% |
| Enterprise tier NPS | +20 → +40 |
| Enterprise churn | 8% → 4% |
Platform Ecosystem
Objective: Build a thriving platform ecosystem
| Key Result | Target |
|---|---|
| Third-party integrations | 50 → 150 |
| Users using integrations | 10% → 30% |
| Partner NPS | +10 → +30 |
Setting OKRs: The Process
1. Start With Company OKRs
What is the company trying to achieve this quarter? Your team OKRs should contribute to company OKRs.
If they don't connect, you're working on the wrong things.
2. Draft Collaboratively
The PM might propose, but the team should input. Engineers often have good ideas about what's achievable and what metrics matter.
Ownership comes from involvement.
3. Calibrate Ambition
- Are these achievable?
- Are they stretch goals?
- Are they impossible?
Get input from people who've done similar work. Check that you have the capacity and resources to pursue these goals.
Tracking OKRs
Weekly Reviews
Review OKR progress weekly. This doesn't mean weekly panic—it means staying aware of trajectory:
- Are you on track?
- Behind?
- What needs to change?
Simple Tracking System
Use red/yellow/green for each KR, with commentary on progress and blockers. Fancy software is optional; consistent review is essential.
Mid-Quarter Check-ins
Assess whether OKRs are still correct. If circumstances changed dramatically, it's okay to adjust.
OKRs are tools, not prisons.
Common OKR Mistakes
Too Many OKRs
Focus is the point. More than 3-5 objectives defeats the purpose. Be ruthless about what makes the cut.
Activity-Based KRs
"Ship feature X" or "Complete Y project" are activities, not outcomes. Always ask: what happens when we do this? The answer is the real KR.
No Baseline
"Improve retention" is meaningless without knowing current retention. Always baseline before setting targets.
Set and Forget
OKRs written in January and reviewed in April are useless. Regular check-ins make OKRs a living tool.
OKRs and Roadmaps
| OKRs | Roadmaps |
|---|---|
| What you want to achieve | How you'll achieve it |
| Outcomes | Outputs |
They should align but remain distinct.
Example
- OKR: "Improve new user activation to 40%"
- Roadmap initiatives: Simplify signup, add onboarding wizard, implement early engagement emails
The initiatives serve the OKR.
When roadmap items don't connect to OKRs, question them. Why are we doing this? If you can't connect work to goals, it might not be worth doing.
Scoring and Retrospectives
End-of-Quarter Scoring
Each KR gets a score, typically 0-1.0 where 0.7 is solid achievement for stretch goals. The Objective's score is an average or judgment across KRs.
Learning Matters More Than Scores
- Why did you achieve or miss?
- What will you do differently?
- What did you learn about your product, users, or capabilities?
Carry Lessons Forward
- If you consistently miss a type of KR, your estimation is off
- If you consistently exceed, you're sandbagging
- Each quarter should improve your OKR quality
Related guides
Get the weekly digest of top product people & jobs
One email a week. No spam.