drafty

OKR template

One objective, three measurable key results — drafted by your agent in seconds. Share the link and get feedback anchored to the exact goal, not buried in a Slack thread two weeks later.

What it is
An OKR (Objectives and Key Results) is a goal-setting structure: one inspiring objective that names the direction, paired with two to four measurable outcomes that tell you whether you got there. OKRs are usually set per quarter, per team.

The format takes five minutes to learn. The real problem is what happens after you draft it: leadership wants changes via email, engineering has questions in Slack, and by week four nobody agrees on which version is current. Draft it with your agent, share the Drafty link, and the comments land on the exact line they're about.

Generate it with your agent

Paste this into Claude, Cursor, or any agent — it drafts the OKR and publishes it as a Drafty link:

claude
Draft a quarterly OKR for {team or project}. Structure it as: 1) Objective — one sentence, inspiring and directional, owned by one team. 2) Key Results — two to four measurable outcomes (not tasks) that confirm the objective was reached. Each key result should have a starting number, a target number, and a due date. 3) A short 'what we're not doing this quarter' note to prevent scope creep. Then publish it to Drafty so I can share a link and collect inline comments — no account needed to reply.

See it on a real one

Live canvas — comment on any elementOpen ↗

What goes in an OKR template

  1. Objective — one sentence. Inspiring, directional, owned by one team. "Become the default tool for design feedback" is an objective. "Improve the product" is not.
  2. Key Results — two to four measurable outcomes. Each needs a number: a baseline, a target, and a timeframe. If there's no number, it's a task, not a key result.
  3. What we're not doing — a short list of things that came up and were deliberately parked. This single section prevents the most expensive mid-quarter conversations.

The mistake most first-time OKR writers make

Key results should measure outcomes, not output. "Launch the onboarding redesign" is a task — it tells you what you shipped, not whether it worked. "Increase 7-day activation rate from 31% to 45% by end of Q3" is a key result — it tells you whether the launch actually mattered.

Most OKRs fail because they smuggle tasks in as key results. The test: if the key result would be marked done the moment the PR merges, it's a task in disguise.

When to write OKRs

OKRs are useful when a team needs to agree on what success looks like before they start working — not just what they'll ship. For a bounded piece of work with a known deliverable, a one-pager or PRD is usually more useful. Reach for OKRs when the goal is a direction, not a feature.

Share the Drafty link with leadership for sign-off and with engineering for scope questions. The comments thread to the exact key result — "is 45% realistic given the current funnel?" lands on that line, not floating in an email.

FAQ

How many key results should an OKR have?

Two to four per objective. Three is the most common. More than four usually means the objective is actually two objectives — split it. Fewer than two usually means the key results aren't specific enough to confirm you reached the goal.

What's the difference between OKRs and KPIs?

KPIs are ongoing health metrics — revenue, uptime, NPS. They never "complete"; you monitor them continuously. OKRs are time-boxed bets: you're saying that this quarter, moving this specific number from here to there is worth the team's focus. A KPI tells you if the business is healthy. An OKR tells you if the bet paid off.

Should key results be tasks or outcomes?

Outcomes. A key result is something you could fail to hit even if you shipped everything on the roadmap. "Launch the feature" can't fail once it ships. "10% more users complete the core action within 7 days" can fail — and that's what makes it a useful signal.

How often should we review OKRs?

Most teams do a quick check-in weekly (5 minutes: is each key result on track, off track, or at risk?) and a fuller review mid-quarter. The mid-quarter review is the one most teams skip — it's also the one that prevents the end-of-quarter surprise where half the key results quietly became impossible in week six.

Can I set OKRs at 100% or should they be stretch goals?

Both approaches work, but they need different scoring. If you set stretch goals (the Google model: 70% hit = success), say so explicitly in the OKR so leadership doesn't interpret a 0.7 score as underperformance. If you set committed goals, aim for 100% — missing them is a signal, not a strategy. The important thing is that everyone reading the OKR knows which type it is.

What's the difference between a company OKR and a team OKR?

Company OKRs set direction for the organization — usually three to five objectives set by leadership. Team OKRs connect to those: each team picks the one or two company OKRs their work most directly moves, then writes their own objectives and key results that cascade from there. If a team's OKRs can't be traced to a company OKR, it's worth asking whether the work is the right priority.