OKR score calculator
Enter each key result's starting point, target and where it actually landed. You get the 0.0–1.0 score for each one and the objective's rollup — including the two cases teams most often get wrong.
- Score0.75
- Score0.67
- Score—
2 of 3 key results scored — an unscored key result counts as 0.0
Between 0.4 and 0.7 is the normal range for ambitious goals — real progress, short of the target.
Nothing is sent anywhere — the arithmetic runs in your browser, and there is no sign-up, no email, and no file to download.
The formula, and the two places it goes wrong
A key result scores itself when it has a baseline and a target. The distance you closed is the score:
score = (result − start) ÷ (target − start), capped at 1.0
Metrics that should go down
Churn from 8% to a 2% target, landing at 4%, scores 0.67 — you closed four points of the six you were after. Dividing the two numbers instead reads as 200%, and a partial win gets filed as a blowout. The formula above already handles this: the denominator changes sign with the direction.
Key results nobody scored
Divide by the number of key results the objective has, not the number that got scored. One at 1.0 alongside two nobody touched is 0.33, not 1.00 — averaging only the scored ones quietly turns an unfinished quarter into a perfect one.
The full method — what to do before the numbers, how to run the grading session, and what belongs on a scorecard — is in OKR grading and scoring.
What to type in each field
Key results come in a few shapes, and the same three numbers cover all of them once you know what counts as a start.
| Shape | Example | Start / Target / Result | Score |
|---|---|---|---|
| Grow a number | Trial signups from 200 to 400 | 200 · 400 · 350 | 0.75 |
| Shrink a number | Monthly churn from 8% to 2% | 8 · 2 · 4 | 0.67 |
| Percentage or rate | Activation from 41% to 60% | 41 · 60 · 52 | 0.58 |
| Countable milestones | 12 help articles rewritten | 0 · 12 · 9 | 0.75 |
| Done or not done | SOC 2 audit passed | 0 · 1 · 1 | 1.00 |
A single-step key result — done or not done — is the one shape worth avoiding where you can. It can only ever score 0 or 1, so it tells you nothing during the quarter and turns the objective into a coin flip at the end. If most of yours look like that, the fix is in how they are written rather than how they are scored: how to write OKRs covers the “from X to Y by [date]” shape, and OKR examples has ready-made ones with real numbers in them.
Reading the number you just got
Strong cycle
You hit, or nearly hit, an ambitious target. Worth saying out loud in the review — and worth checking that it was a stretch, because a steady 1.0 every quarter usually means the bar was set where you already knew you would land.
The normal range
Real progress, short of the target. This is where most honest OKRs land, and it is not a failure — the question for the review is which part moved and which part stalled, not who is to blame.
Something was off
Either the target was unrealistic, the work was blocked all quarter, or nobody updated the numbers. All three are worth knowing, and only the first two are about the goal.
Two things the number cannot tell you, and both decide whether it means anything. Whether the target was a genuine stretch when it was set — sandbagged targets score beautifully — and whether scores are tied to anyone’s pay, because the moment they are, next quarter’s targets get easier. Both are on the list of common OKR mistakes, and both are cheaper to avoid than to unwind.
If you run SMART goals, this scale isn’t yours
The 0.0–1.0 scale exists because OKRs are meant to be a stretch you probably won’t fully reach. A SMART goal is the opposite bargain: it is written to be achievable, so scoring it 0.7 would mean it failed. Close those with met, partial or missed instead. If you are not sure which one you are running, OKR vs SMART goals lays out the difference, and the OKR guide is the place to start if the framework itself is new.
Scoring questions
How do you calculate an OKR score?+
Score each key result from 0.0 to 1.0 with (result − start) ÷ (target − start), capped at 1.0. The objective's score is the average of its key results. Around 0.7 is the sweet spot on a genuine stretch — a steady 1.0 usually means the goal was too safe.
How do you score a key result where the metric should go down?+
The same formula, unchanged — measure the distance you closed, not the ratio of the two numbers. Churn from 8% to a 2% target that landed at 4% scores (4 − 8) ÷ (2 − 8) = 0.67. Dividing 8 by 4 instead reads as 200% and turns a partial win into a blowout.
What happens to a key result nobody scored?+
It counts as 0.0 in the objective's average, because the average divides by the number of key results the objective has — not the number that got scored. One key result at 1.0 next to two nobody touched is 0.33, not 1.00.
Is a score of 1.0 good?+
Once, maybe. Every cycle, no — it usually means the targets were set where you already knew you would land. OKRs are meant to be a stretch, which is why roughly 0.7 is the target rather than the full mark.
Or let the scoring happen as you go.
Celorly keeps the baseline, the target and the result on the goal itself — and rolls the objective up for you, with the same rule this page uses.