Most teams write OKRs full of lagging indicators — final outcomes visible only at quarter's end, too late to act on. The three-type system adds leading signals and guardrails.
▸ Try the interactive toolModule 1 introduced OKRs as outcome-based goal setting. This goes deeper: most teams write OKRs containing only lagging indicators — final outcomes visible only at the end of the quarter, far too late to course-correct. The three-type Key Result system adds leading KRs (early-warning signals) and guardrail KRs (protection against doing harm while chasing the goal).
A lagging KR tells you whether you succeeded after it's too late to change anything. A leading KR is an earlier signal that predicts the lagging one — so you can see trouble coming and act. A guardrail KR protects against the classic failure of hitting a target while breaking something else (boosting signups while tanking retention). Together, the three types make an OKR not just a scoreboard but a steering instrument you can act on mid-quarter.
Lagging — the final outcome (e.g. quarterly revenue) — true but late
Leading — an early signal that predicts the lagging KR — actionable in time
Guardrail — a metric you must not harm while pursuing the objective
An OKR built only on lagging indicators is a rear-view mirror — it tells you what happened when it's far too late to influence it.
| The shortcut | What it costs | What it gives you instead |
|---|---|---|
| Only lagging KRs | You learn you missed the goal at quarter-end, too late to act. | Leading KRs give early signals you can still respond to. |
| Gaming the objective | Hitting the KR while breaking something else (e.g. retention). | Guardrail KRs protect against winning the battle, losing the war. |
| No mid-quarter steering | Nothing actionable until the final number lands. | Leading indicators turn OKRs into a steering instrument. |
| Tunnel vision on one metric | Optimising a single number distorts the wider system. | Guardrails keep the broader health of the system in view. |
Start with the genuine outcome you want — the final, measurable result. This is the destination, even though it's only visible at the end.
Identify an earlier metric that reliably moves before the lagging one. If the lagging KR is quarter-end retention, a leading KR might be early-week engagement — visible in time to act.
Name the thing you must not damage while chasing the objective. Pushing growth? Guardrail on retention or quality. This prevents a hollow win.
A leading KR only helps if it actually predicts the lagging one. Validate the relationship with data, or you're steering by a false signal.
The whole point is early action. Watch leading indicators through the quarter and adjust — don't wait for the lagging number to confirm a miss you could have prevented.
A team set an OKR with a single lagging key result: end-of-quarter revenue retention. They executed all quarter with no early read, and only at quarter-end discovered they'd missed — with no time left to do anything about it.
Redesigned with three types, the same objective gained a leading KR (early-week engagement, which reliably predicted retention) and a guardrail (support quality, so they wouldn't chase retention by over-promising). Mid-quarter, the leading indicator dipped — and because they saw it early, they corrected course and recovered the lagging outcome instead of just reporting the miss.
The deliverable is one objective with KRs of all three types — a lagging outcome, a leading predictor, and a guardrail.
| KR type | Answers | Example role |
|---|---|---|
| Lagging | Did we succeed? | The real outcome — but visible late |
| Leading | Are we on track now? | Early signal you can act on |
| Guardrail | Are we breaking anything? | A metric you must not harm |
A goal you can only measure at the finish line is a goal you can't steer toward. Leading indicators turn OKRs from a verdict into a control system — and guardrails stop you from winning the metric while losing the business.
The hard part is the leading–lagging link: a leading KR is only useful if it genuinely predicts the outcome, and teams often pick an early metric that's easy to measure but doesn't actually move the lagging one. Validating that relationship is what separates real steering from false comfort. Done well, the three-type system changes how a quarter feels — instead of executing blind and learning the result too late, the team watches leading signals, respects guardrails, and corrects course while it still can.
A rear-view-mirror OKR. Add leading indicators you can act on mid-quarter.
An early metric that doesn't actually predict the outcome is false comfort. Verify the link.
Chasing one number unprotected invites gaming and collateral damage. Add a guardrail.
If you don't review them mid-quarter, leading KRs are pointless. Watch and act on them.
This builds on the outcome-based OKR basics (Module 1, Tool 20) with the three-type refinement.
Leading and lagging KRs often decompose the North Star metric into early and final signals (Module 1 / Module 5).
OKRs set the outcomes; the roadmap (Tool 25) sets the work meant to move them.
Validating the leading–lagging link is an analytics exercise — the heart of Module 5.
Take an OKR with a single end-of-period outcome (a lagging KR). Brainstorm one earlier metric that would predict it — something you'd see moving weeks before the final number.
Then add a guardrail: what must you not harm while chasing this objective?
If you can name a leading indicator you'd genuinely act on mid-quarter, you've turned a scoreboard into a steering wheel — the entire point of advanced OKR design.
New to this? Start here · OKRs →