A goal-setting framework that pairs ambitious Objectives (what you want to achieve) with measurable Key Results (how you'll know you've achieved it).
Key Takeaways
OKRs (Objectives and Key Results) is a goal-setting framework that helps organizations define what they want to accomplish and measure whether they're actually getting there. An Objective is a clear, qualitative statement of where you want to go. It should be ambitious, time-bound, and inspiring enough that people genuinely care about achieving it. Key Results are the measurable outcomes that tell you whether you've reached the Objective. They're specific, quantifiable, and verifiable. You either hit the number or you didn't. There's no room for "sort of" with a well-written Key Result. The beauty of OKRs is their simplicity. One Objective. Two to five Key Results. That's it. The discipline isn't in the format. It's in the choices you make about what to focus on and, just as importantly, what to leave off the list.
Andy Grove developed the OKR framework at Intel in the 1970s, building on Peter Drucker's Management by Objectives (MBO) concept. Grove stripped away the bureaucracy of MBOs and created something leaner and faster. John Doerr, a former Intel employee turned venture capitalist, brought OKRs to Google in 1999 when the company had about 40 people. He literally walked into a conference room and taught Larry Page and Sergey Brin the framework. Google's explosive growth and very public credit to OKRs turned the system into a standard across Silicon Valley. Today, companies of every size and industry use OKRs. Intel, Google, LinkedIn, Spotify, Twitter, The Gates Foundation, and thousands of mid-market companies all run on some version of the framework. What started as one manager's approach to goal-setting at a chip company became the default operating system for organizational focus.
An Objective answers the question: "Where do we want to go?" It should be inspiring and clear enough that anyone in the company can understand it without a glossary. "Become the most trusted brand in our category" is a good Objective. "Optimize multi-channel brand equity metrics" is not. Key Results answer the question: "How will we know we got there?" Each one should be a number: a percentage, a dollar amount, a count, or a deadline tied to a concrete deliverable. "Increase NPS from 42 to 65" is a strong Key Result. "Improve customer satisfaction" is not, because there's no way to score it objectively at the end of the quarter. Most companies set OKRs quarterly at the team level and annually at the company level. Progress gets tracked weekly or biweekly in team check-ins. At the end of each quarter, teams score their Key Results on a 0.0 to 1.0 scale, run a retrospective, and set new OKRs for the next cycle. The rhythm of set, track, score, reflect is what makes OKRs more than just a fancy to-do list.
Setting good OKRs isn't complicated, but it does require discipline. Most teams that struggle with OKRs don't have a framework problem. They have a writing problem. Follow these five steps and you'll avoid the most common pitfalls.
Before any team writes OKRs, leadership needs to set three to five company-level Objectives for the quarter. These serve as the north star that everything else aligns to. Every team OKR should connect back to at least one company Objective. Without this top-down alignment, you end up with teams working hard on things that don't actually move the business forward. The executive team should finalize company OKRs at least a week before teams start drafting theirs, so there's time for people to understand the priorities and ask questions.
Each team writes one to three Objectives that describe their most important contributions to the company priorities for the quarter. Use plain language. "Reduce customer churn to protect revenue growth" is better than "Optimize retention metrics across the customer lifecycle." Every Objective should pass the "so what?" test. If nobody would notice whether you achieved it, it's not worth setting. Objectives should be ambitious but not delusional. They should make the team stretch without making them roll their eyes.
For each Objective, define two to five Key Results that are specific and quantifiable. Good Key Results start with a verb and include a number. "Reduce average support ticket resolution time from 48 hours to 24 hours" is a strong Key Result. "Improve customer support speed" is not, because there's no way to score it objectively at the end of the quarter. Each Key Result should have a clear starting point and a clear target. If two reasonable people would disagree about whether you hit it, rewrite it until they wouldn't.
Decide whether each OKR is a "committed" OKR (you expect to hit 100%) or an "aspirational" OKR (70% completion counts as a win). Google popularized the 70% standard for aspirational OKRs. The logic is straightforward: if you're hitting 100% of your stretch goals every quarter, you're not stretching enough. If you're hitting 30%, your goals are disconnected from reality. Label each OKR clearly so everyone knows the expectation. Mixing committed and aspirational OKRs without labeling them leads to confusion about what's actually expected.
Share draft OKRs across teams before finalizing them. Look for three things: conflicts (two teams pulling in opposite directions), dependencies (one team's Key Result requires another team's output), and gaps (important company priorities that no team has picked up). Fix these before locking in the quarter. Then publish final OKRs where everyone in the company can see them. Transparency is a core principle of the OKR framework. When everyone can see everyone else's goals, alignment happens naturally and cross-team collaboration becomes much easier.
OKRs, KPIs, MBOs, and SMART goals are all related to goal-setting, but they serve different purposes and work differently in practice. Here's how to tell them apart and when to use each one.
| Dimension | OKRs | KPIs | MBOs | SMART Goals |
|---|---|---|---|---|
| What it is | Goal-setting framework with ambitious Objectives and measurable Key Results | Ongoing performance metrics that track health of a process or outcome | Goal-setting method where managers and employees agree on objectives tied to evaluation | Individual goal format: Specific, Measurable, Achievable, Relevant, Time-bound |
| Purpose | Drive focus, alignment, and stretch performance across teams | Monitor ongoing performance against established targets | Set individual goals and evaluate performance against them | Create well-defined personal or project goals |
| Cadence | Quarterly (most common), with weekly check-ins | Continuous, always-on monitoring | Annual, tied to yearly performance review cycle | Varies: per project, per quarter, or per year |
| Ambition level | Deliberately stretch. 70% completion is healthy. | Expected to be met or exceeded. Missing a KPI is a problem. | Set to be achievable. Hitting 100% is expected. | Achievable by definition. The A in SMART stands for it. |
| Tied to compensation | No. Linking to bonuses kills ambitious goal-setting. | Sometimes. Bonus structures may reference KPI targets. | Yes. MBOs are typically tied to annual reviews and bonuses. | Varies. Often used informally without compensation link. |
| Transparency | Public across the organization | Varies. Often department-level. | Private between manager and employee. | Usually private or shared with direct manager only. |
| Best for | Company and team alignment, driving ambitious outcomes | Monitoring business health and operational performance | Individual accountability in traditional management structures | Clarifying personal goals or project deliverables |
Seeing real examples makes the OKR format click faster than any explanation. Here are sample OKRs for six common departments. Adapt the numbers to your own baselines and targets.
| Department | Objective | Key Result 1 | Key Result 2 | Key Result 3 |
|---|---|---|---|---|
| Engineering | Ship a product customers love using every day | Increase daily active users from 12K to 20K | Reduce P1 bug count from 15 to fewer than 5 per month | Achieve 99.95% uptime (up from 99.8%) |
| Sales | Accelerate revenue growth in Q3 | Close $2.4M in new business (up from $1.8M) | Increase average deal size from $18K to $24K | Shorten sales cycle from 62 days to 45 days |
| Marketing | Make our brand the first name prospects think of | Grow organic search traffic from 80K to 130K monthly visits | Generate 1,200 marketing-qualified leads (up from 800) | Increase brand awareness survey score from 22% to 35% |
| HR / People | Build a hiring engine that attracts top talent | Reduce time-to-hire from 45 days to 28 days | Increase offer acceptance rate from 72% to 85% | Achieve 4.5+ candidate experience score (out of 5) |
| Finance | Build a financial foundation for sustainable growth | Reduce monthly close from 12 days to 5 days | Improve cash flow forecast accuracy to within 5% variance | Cut accounts receivable over 90 days from $400K to $150K |
| Customer Success | Turn customers into advocates | Improve NPS from 42 to 60 | Reduce churn rate from 5.2% to 3.0% | Increase expansion revenue by 30% |
Most OKR implementations fail not because of the framework itself, but because of how teams apply it. The framework is simple. The execution is where things go wrong. These are the five mistakes that kill OKR programs most often.
The most frequent mistake is writing Key Results that describe activities rather than outcomes. "Launch a new email campaign" is a task. "Increase email-generated revenue from $50K to $80K per month" is an outcome. The difference matters because you can complete a task without creating any business impact. If your Key Result can be checked off without anything actually improving, it's a task disguised as a Key Result. Always ask: "What change will we see if this works?" That's your real Key Result.
If everything is a priority, nothing is. Teams with ten Objectives and forty Key Results don't have focus. They have a to-do list formatted as OKRs. The whole point of the framework is forcing prioritization. Limit each team to three to five OKRs per quarter with two to five Key Results each. If you can't narrow down, ask the team: "If we could only accomplish one thing this quarter, what would matter most?" Start there and add only what's truly essential.
The moment you tie OKR completion to bonuses or performance ratings, people stop setting ambitious goals. They'll sandbag targets to guarantee 100% completion. OKRs are designed to encourage stretch thinking. If hitting 70% of a stretch goal gets you punished financially, the system breaks. Use OKRs for alignment and direction. Use separate KPIs and performance metrics for compensation decisions.
OKRs that get written in January and reviewed in March are just documentation. The framework only works with regular check-ins: weekly team updates on Key Result progress, biweekly one-on-ones discussing blockers, and a formal midpoint review at the halfway mark of each cycle. If you're not talking about OKRs every week, you're not using them.
When leadership dictates all OKRs without team input, you lose the buy-in that makes the framework effective. People don't commit to goals that were handed to them. They commit to goals they helped create. The best practice is a mix: company-level Objectives flow from the top, but teams write their own OKRs that align upward. This creates genuine ownership. Research consistently shows that employees who participate in setting their goals are significantly more committed to achieving them than those who receive goals from above.
These five practices separate teams that get real value from OKRs from teams that treat them as a quarterly paperwork exercise.
OKRs should be understandable by anyone in the company without context. If a new employee can't read your OKR and immediately understand what success looks like, rewrite it. Avoid jargon, acronyms (except widely known ones), and corporate-speak. "Become the #1 rated product on G2" is clear. "Optimize market positioning across key review platforms" is not. The best OKRs read like plain English. They don't need a decoder ring.
At the end of each quarter, score every Key Result on a 0.0 to 1.0 scale. Be honest. A team that consistently scores 1.0 on every Key Result isn't performing well. They're setting easy goals. A healthy pattern is a mix of 0.6 to 0.8 scores on aspirational OKRs and 1.0 on committed OKRs. Treat the scoring conversation as a learning opportunity, not a judgment.
After scoring, hold a 30-minute retrospective. What did we learn? Which Key Results were poorly written? Which Objectives turned out to be less important than we thought? What should we do differently next quarter? This reflection loop is how OKR practice improves over time. Teams that skip it make the same mistakes quarter after quarter.
Publish OKRs where everyone can see them. This could be a shared Notion page, a dedicated Slack channel, an OKR tool like Lattice or Quantive, or even a physical board in the office. Transparency serves two purposes: it creates accountability, and it helps teams spot alignment issues or collaboration opportunities they might otherwise miss.
If your organization has never used OKRs, don't roll them out to every team simultaneously. Pick one team (ideally one that's enthusiastic about trying it), run two or three quarterly cycles, learn what works, and then expand gradually. Most company-wide OKR launches that happen all at once create confusion and frustration because there's no internal expertise to guide teams through the inevitable learning curve. Build that expertise on a small team first, then use those people as coaches when you expand.
These numbers capture the current state of OKR adoption and outcomes across organizations of all sizes.