Sign inGet started
Skills / Productivity-habits

OKRs (Objectives and Key Results)

Where do I want to go? The answer to that is the objective. How will I know I'm getting there? That gives us the key results.

The OKR framework, invented by Andy Grove at Intel in the late 1970s, gives individuals and teams a precise two-part structure for goal-setting: a qualitative, inspirational Objective paired with 2–5 quantitative Key Results that prove whether you got there. Unlike vague goal-setting, OKRs enforce a hard distinction between aspiration and measurable outcome, graded on a 0.0–1.0 scale at the end of every cycle. Used at Intel, Google, Amazon, and thousands of organizations worldwide, OKRs are the gold standard for aligning ambition with accountability.

By Andy Grove · Free
Specimen 01 · Live diagnosisOKRs (Objectives and Key Results)
Input

“For my product team this quarter. Our focus is improving user retention, but I'm not sure how to structure it. Here's what I have so far: 'Objective:…”

Diagnosis
I work with Andy Grove's OKR framework — every goal we write will follow the same formula: one qualitative Objective (the 'what' and 'why') paired with 2–4 Key Results that are numeric and measurable.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

If it doesn't have a number, it isn't a Key Result

The OKR framework is built on a strict two-part formula: an Objective, which must be qualitative and aspirational (a verb-driven statement of what you want to achieve), paired with 2–5 Key Results, each expressed as a specific numeric target that proves the Objective was met. OKRs are set on quarterly or annual cycles and graded 0.0–1.0 at cycle end — Grove's documented 'sweet spot' is 0.6–0.7, signaling the goal was ambitious enough that 1.0 would have been too easy. The framework distinguishes Committed OKRs (operational targets where 1.0 is expected) from Aspirational/Moonshot OKRs (stretch goals where 0.7 is a win). OKRs cascade from company → team → individual, with Grove and Doerr recommending that roughly 40–60% of OKRs originate bottom-up from employees. A central diagnostic runs through the entire framework: Key Results must measure outcomes, not activities — 'Launch the feature' is a task; 'Increase feature adoption from 12% to 40%' is a Key Result.

The problem

Most goal-setting fails because people confuse tasks with outcomes, and ambition with accountability. Teams write to-do lists disguised as goals ('Launch the new dashboard', 'Improve customer experience') with no way to know at cycle end whether they succeeded. Grove identified the root failure: goals that don't separate the inspirational 'what' from the measurable 'how will we know' collapse into activity tracking — everyone stays busy, but no one can prove the work mattered.

The solution

Walk away with a complete, Grove-compliant OKR set — Objectives that inspire, Key Results that measure outcomes not activities, and a grading structure that forces honest reckoning at the end of every cycle.

You bring
  • A goal, initiative, strategy, or area of focus you want to set OKRs around
  • The team or individual level these OKRs are for (company, team, or individual contributor)
  • Any existing draft OKRs or task lists you want converted or reviewed
  • Your time horizon (quarterly or annual cycle)
You get
  • A complete OKR set (3–5 Objectives, each with 2–4 measurable Key Results) grounded in Grove's formula
  • A KR Diagnostic Report flagging activity-based KRs, vanity metrics, and missing numeric targets with specific rewrites
  • Each OKR labeled as Committed or Aspirational with documented grade expectations (0.7 vs. 1.0 target)
  • An end-of-cycle grade sheet (0.0–1.0 per KR) with retrospective commentary on what the scores signal
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“Your Q2 planning doc has six 'goals' including 'Improve customer satisfaction,' 'Move faster as a team,' and 'Launch the new dashboard.' At the end of the quarter you can't tell whether you succeeded — satisfaction is unmeasured, speed is undefined, and the dashboard launched but nobody is using it.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With OKRs (Objectives and Key Results)

“Your Q2 OKR sheet has three Objectives, each with 3 Key Results carrying numeric baselines and targets. On June 30 you run the numbers: KR grades of 0.8, 0.6, 0.4, 0.9, 0.7, 0.7, 0.5, 0.8, 0.6 — Objective averages of 0.63, 0.72, and 0.63. You know exactly where you hit your stretch goals, where you fell short, and whether your ambition was calibrated correctly for next quarter.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
OKRs (Objectives and Key Results)Claude
OKRs (Objectives and Key Results)
I work with Andy Grove's OKR framework — every goal we write will follow the same formula: one qualitative Objective (the 'what' and 'why') paired with 2–4 Key Results that are numeric and measurable. At the end of a cycle, each KR gets a grade from 0.0 to 1.0. Grove's sweet spot is 0.6–0.7 — if you're hitting 1.0 every time, your bar is too low. What are we setting OKRs for — a team, an individual, or a company level?
For my product team this quarter. Our focus is improving user retention, but I'm not sure how to structure it. Here's what I have so far: 'Objective: Improve retention. KR1: Send better onboarding emails. KR2: Fix the bugs causing drop-off. KR3: Increase retention.'
Good start — let me run a diagnosis before we rewrite. You have three problems here:
KR1 ('Send better onboarding emails') — this is a task, not an outcome. It tells me what you'll do, not what will change. Grove's test: can you assign it a 0.0–1.0 grade at quarter end? You can't grade 'better.'
KR2 ('Fix the bugs causing drop-off') — same issue, plus 'fix' is binary and vague. Which bugs? How many? What's the baseline drop-off rate?
KR3 ('Increase retention') — this is an Objective disguised as a KR. Increase from what to what? By when?
Now let me ask: what does retention actually mean for your product? Day-7 retention, Day-30, monthly active users coming back — which metric do you track?
We track Day-30 retention. It's currently at 34%. We also watch the onboarding completion rate — right now only 41% of new users finish the onboarding flow.
Perfect — now we have baselines, which means we can write real Key Results. Here's a rewrite:
Objective: Make every new user feel the product's value before their first week ends(Qualitative, aspirational, time-implied — this is the 'why' that motivates the work)
KR1: Increase Day-30 retention from 34% to 46%KR2: Increase onboarding flow completion from 41% to 65%KR3: Reduce the bug-related drop-off events in the first 72 hours from 1,200/week to under 400/week
Each KR is now falsifiable — at the end of Q2 you run the numbers and give each one a grade from 0.0 to 1.0. I'd classify all three as Aspirational OKRs — these are meaningful stretches, not guaranteed operational targets. A 0.7 average at cycle end would be a strong quarter. Does this match the ambition level you're aiming for, or should any of these be more conservative (Committed)?
The Day-30 retention one feels right as a stretch. But KR2 — onboarding completion — we've already started the fix, so we're pretty confident about that one. Should it be a different type?
Yes, exactly — that's when you flip it to a Committed OKR. If your team is already mid-implementation and 65% is achievable with solid execution, then 65% is the floor, not the moonshot. Grade expectation is 1.0; missing it is a problem. You might raise the target (say, 72%) to keep it honest as an Aspirational, or lock in 65% as Committed and let the other two KRs carry the stretch ambition. Grove's rule: a mix of Committed and Aspirational OKRs in the same Objective is perfectly valid — just be explicit about which is which so you're not sandbagging on paper while calling it a stretch.
That makes sense. I'll keep 65% as Committed and push Day-30 retention and the bug KR as Aspirational. This is way cleaner than what I had.
Reply to OKRs (Objectives and Key Results)
What the skill is doing
01OKR Drafting Workshop
02KR Anti-Pattern Diagnosis
03Committed vs. Aspirational Classification
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
O(OKRs (Objectives and Key Results)typing…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
For my product team this quarter. Our focus is improving user retention, but I'm not sure how to structure it. Here's what I have so far: 'Objective:…9:14✓✓
At the end of a cycle, each KR gets a grade from 0.0 to 1.0.9:14
Grove's sweet spot is 0.6–0.7 — if you're hitting 1.0 every time, your bar is too low.9:15
What are we setting OKRs for — a team, an individual, or a company level?9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

OKR Drafting Workshop

Takes a goal, project, or strategic initiative and structures it into a proper Objective (qualitative, verb-driven, inspirational) paired with 2–4 Key Results (each with a specific numeric baseline and target). The output follows Grove's exact formula — no KR without a number, no Objective that is itself a metric.

Based on Grove's two-question structure from 'High Output Management': 'Where do I want to go?' (the Objective) and 'How will I know I'm getting there?' (the Key Results), with the rule that KRs must be falsifiable — a yes or no at cycle end.
CapabilityC-02

KR Anti-Pattern Diagnosis

Reviews existing OKRs and flags specific violations: activity-based KRs that measure outputs rather than outcomes, vanity metrics (impressions, downloads without engagement signals), Objectives that are themselves numbers, and KRs with no defined baseline. Each flag comes with a specific rewrite.

Applies Grove's documented anti-pattern library — the distinction between 'Launch the feature' (an output/task) vs. 'Increase feature adoption from 12% to 40%' (an outcome), and John Doerr's extension identifying vanity metrics as a common KR failure mode.
CapabilityC-03

Committed vs. Aspirational Classification

Labels each OKR as Committed (operational target, 1.0 expected at cycle end) or Aspirational/Moonshot (stretch goal, 0.7 is a win), and documents what each grade means in context. Prevents the common mistake of treating all OKRs the same, which causes either sandbagging or demoralization.

Reflects the Committed vs. Aspirational OKR distinction formalized by Doerr in 'Measure What Matters' and present in Google's re:Work OKR guide, rooted in Grove's insight that the 0.6–0.7 average grade signals appropriate ambition — 1.0 means the bar was set too low.
CapabilityC-04

Cascade Alignment Builder

Takes company-level or team-level OKRs and drafts the next level down — translating a company Objective into team-level OKRs, or a team Objective into individual contributor OKRs that roll up meaningfully. Flags misalignments where individual OKRs don't connect to team priorities.

Implements Grove's cascading alignment model — OKRs flow company → team → individual — with the Doerr/Grove recommendation that 40–60% of OKRs originate bottom-up to ensure ownership alongside strategic alignment.
CapabilityC-05

End-of-Cycle Scoring & Retrospective

At the end of a quarterly or annual cycle, scores each Key Result on Grove's 0.0–1.0 scale, computes the Objective average, and provides retrospective commentary: what the grades signal about ambition calibration, which KRs were systematically avoided, and what to carry forward or retire.

Applies Grove's grading rubric (0.0 = not started, 0.3 = started but far short, 0.7 = ambitious success, 1.0 = fully achieved or too easy) with the documented sweet-spot principle that a team averaging 0.6–0.7 set the right level of stretch.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

OKR Cycle Sheet

A structured document for the full cycle: each Objective listed with its 2–4 Key Results, their baseline and target values, their Committed/Aspirational classification, and end-of-cycle grade fields (0.0–1.0 per KR plus Objective average). The canonical auditable artifact Grove's framework produces.

OutputD-02

KR Diagnostic Report

A flagged review of existing OKRs listing every anti-pattern found — activity-based KRs, missing numeric targets, vanity metrics, Objectives that are numbers — each with a specific rewrite that transforms the violation into a Grove-compliant Key Result.

OutputD-03

Cascade Alignment Map

A hierarchical view showing how company-level Objectives connect to team-level OKRs and, where applicable, individual contributor OKRs — with explicit linkage lines showing which KRs roll up into which higher-level Objective.

OutputD-04

Quarterly OKR Planning Template

A pre-structured planning document for the next cycle, pre-populated with example Objectives and Key Results for the user's context, ready to fill in with real targets, baselines, and ownership assignments.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Andy Grove

Andy Grove invented the OKR framework at Intel in the late 1970s and first documented it in 'High Output Management' (1983), now considered one of the most influential management books ever written. As CEO of Intel from 1987 to 1998, he used OKRs to guide one of Silicon Valley's most consequential strategic pivots — from memory chips to microprocessors — growing the company into a global powerhouse. Grove's framework was later brought to Google by early Intel employee John Doerr, seeding its adoption across Amazon, LinkedIn, and hundreds of major organizations worldwide.

Status · Inspired by Andy Grove’s work — not yet claimed. Are you Andy Grove?
Primary sourceS-01

High Output Management

by Andy Grove

CEO of Intel (1987–1998); inventor of OKR framework; author of 'High Output Management' (1983) and 'Only the Paranoid Survive' (1996); Time Magazine Man of the Year 1997.

Read the original ↗
Citationrework.withgoogle.com
In the build queue

Be first to run it.

OKRs (Objectives and Key Results) is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail
At launchI want to set OKRs for my team this quarter — our focus is [describe your area, e.g., 'growing enterprise revenue' or 'reducing churn']. Can you help me draft them and check if my Key Results are actually measurable outcomes or just tasks?