A key result is a specific, measurable outcome that tells you whether you’re making progress toward a larger goal. It’s one half of the OKR (Objectives and Key Results) framework, a goal-setting system used by teams and organizations to define what they want to achieve and how they’ll know they got there. The objective is the qualitative destination; the key result is the quantitative proof you arrived.
How Key Results Fit Into OKRs
An OKR pairs one objective with a small set of key results, typically two to five. The objective describes what you want to accomplish in plain, motivating language. Key results then answer the follow-up question: how will we measure success?
Each key result needs to be specific, time-bound, and verifiable. There should be no gray area about whether you hit the mark. You either met the metric or you didn’t. Once all the key results under an objective are completed, the objective itself is considered achieved. If only some are met, the objective was only partially reached, and the scoring system (more on that below) captures exactly how close you got.
Outcomes, Not Tasks
The most common mistake people make when writing key results is listing activities instead of outcomes. A key result should describe a change in a metric, not a to-do item. The difference matters because completing a task doesn’t guarantee the result you actually care about.
Here’s a concrete example. Say your objective is “reduce the number of data errors in the system.” A bad key result would be “install release 10.0 of the vendor package.” That’s a task. You can check the box without knowing whether errors actually went down. A good set of key results would measure the thing you’re trying to change: the number of data quality errors reported to the support desk, the number of orders that can’t be filled automatically, and the order errors reported by customers. Those are outcomes you can track over time.
A useful test: if you can finish your key result in a single action, it’s probably a task. If it requires sustained effort and shows up as a number moving in the right direction, it’s likely a real key result. Look for leading indicators, metrics you can check regularly and that predict progress toward the objective, rather than lagging indicators you can only measure after the fact.
What Good Key Results Look Like
Strong key results share a few traits. They start with a number or end with one. They specify a direction (increase, reduce, reach). And they’re ambitious enough to require real effort but realistic enough that the team believes the goal is possible within the time frame.
Some examples across different functions:
- Sales objective: grow revenue from enterprise accounts. Key results might include increasing average deal size from $40,000 to $55,000, closing 15 new enterprise contracts, and reducing the average sales cycle from 90 days to 65 days.
- Product objective: improve the new-user onboarding experience. Key results could be raising the percentage of users who complete onboarding from 45% to 70%, reducing time-to-first-action from 8 minutes to 3 minutes, and increasing the 7-day retention rate from 30% to 50%.
- Customer support objective: deliver faster, higher-quality responses. Key results might target reducing median first-response time from 4 hours to 1 hour, raising customer satisfaction scores from 3.8 to 4.5, and keeping the ticket reopen rate below 5%.
Notice that none of these key results describe how the team will get there. The “how” lives in the initiatives, projects, and daily work that the team decides on separately. Key results define the finish line; the team picks the route.
Committed vs. Aspirational Key Results
Not all key results carry the same expectations. Organizations that use OKRs often distinguish between two types: committed and aspirational.
Committed key results are goals the team is expected to hit in full and on time. They’re ambitious but realistic. Grading is binary: you either achieve 100% or you don’t. If a team realizes mid-cycle that a committed key result is at risk, the expectation is to escalate quickly and find a path to delivery.
Aspirational key results, sometimes called moonshots or 10x goals, are deliberately set beyond what a team can fully accomplish in a single cycle. They exist to push thinking further than a committed goal would. Falling short is expected. Google, which popularized the OKR framework internally, expects teams to hit roughly 70% of their aspirational key results on average. An aspirational key result that stays unfinished doesn’t get deleted. It remains on the list until it’s eventually completed, sometimes carrying over across multiple quarters.
The distinction matters because it changes how you interpret the score. A committed key result at 60% is a miss. An aspirational key result at 60% represents solid progress.
How Key Results Are Scored
At the end of a cycle (usually a quarter), each key result gets a score. A widely used approach is a 0.0 to 1.0 decimal scale, where 1.0 means fully achieved. To get the overall score for an objective, you average the scores of its individual key results, with each one weighted equally.
For aspirational key results, the scoring bands typically look like this:
- 0.7 to 1.0 (green): On track. What the team is doing is working.
- 0.4 to 0.6 (yellow): At risk. The approach may need adjusting.
- 0.0 to 0.3 (red): The key result likely won’t be met. Time to reassess the strategy and get alignment on next steps.
For committed key results, scoring is simpler: pass or fail. The target is 1.0. Anything less is a miss.
Scores alone don’t tell the full story, though. Most organizations pair numerical grading with a self-assessment, where the team reflects on what worked, what didn’t, and what external factors affected the outcome. A key result that scored 0.5 because the market shifted mid-quarter tells a very different story than one that scored 0.5 because the team underestimated the effort. The number starts the conversation; the reflection makes it useful.
Writing Better Key Results
If you’re setting key results for the first time, a few practical guidelines help. Start by asking what would have to be true for the objective to be achieved. Your answers are your candidate key results. Then pressure-test each one: can you attach a number to it? Can you measure it before the cycle ends? Would two people looking at the same data agree on the score?
Keep the set small. Two to five key results per objective is the standard range. More than that usually means the objective is too broad and should be split. Each key result should measure a different dimension of success rather than restating the same metric in slightly different ways. For a revenue objective, you might pair a growth metric with a retention metric and an efficiency metric, covering breadth rather than redundancy.
Finally, key results should evolve. They’re written at the start of a cycle based on what you know at the time, but as work progresses and you learn more, adjusting a key result to reflect new information is better than rigidly chasing a metric that no longer matters.