Stretch goals, without the damage.
Stretch targets exist to change how a team thinks, not just how hard it works. Used well they produce step changes. Unlabelled, they quietly poison the whole goal system.
An aspirational target deliberately raises the bar past what a plan can promise, because the plan you can promise is, by definition, the thinking you already had. "From 31% to 38%" gets you optimisation. "From 31% to 55% (aspirational)" forces the better question: what would have to be true for that to happen? Different number, different conversations, different quarter.
The contract that makes this safe: 100% achievement is not expected, and success is judged in the retrospective on graded outcomes, learning, effort and collaboration included, rather than pass/fail. Fifty percent progress on a genuine stretch can be an excellent quarter. That sentence has to be true in your organisation, out loud, before stretch targets are usable at all.
The one non-negotiable: the label
Always write it down: "Increase [metric] from X to Y (aspirational)". The label is not decoration, it is the difference between a stretch goal and a lie. An unlabelled stretch target reads as a commitment, and people respond to missed commitments the way people always do: they sandbag the next target, they stop surfacing risk, and within two quarters your goal system is a negotiation. Every pathology attributed to "OKRs not working" that we are shown, more than half trace back to unlabelled stretch.
Two honest tensions
Our methodology holds two pressures against stretch targets, deliberately, because both are right.
The first comes from the moonshot side: decomposing a grand ambition into quarterly targets can degrade the ambition while keeping the poster. The ten-year aspiration does not survive being chopped into twelve-week instalments unless each instalment stays tethered to the underlying logic, the what would have to be true discipline.
The second comes from Rumelt, and it pushes the other way entirely: proximate objectives argues that good leadership sets targets close enough to hit, with the leader absorbing the ambiguity rather than delegating it as "ambition". This is a genuine tension, not a contradiction to be smoothed over: stretch works when a team has the capability and the safety to think bigger, proximate works when ambiguity is drowning them. Choosing which situation you are in is a leadership judgement, and pretending the tension does not exist is how frameworks replace thinking.
Money is the other killer
Attach cash to a stretch goal and you have not created ambition, you have created a floor. The moment a bonus depends on 70%, 70% becomes the target, and every behaviour the stretch was meant to unlock, risk-taking, honest confidence scores, big swings, inverts. Aspiration has to be protected from the bonus plan by design, which is why compensation connects to OKRs only through a deliberately insulated route in our methodology, the SHOP model.
In practice
Set stretch targets in the open, label them in the text of the Key Result itself, and grade them in the retrospective with the learning on the table. And watch for the tell in weekly check-ins: a team that stops reporting Low confidence on a stretch target has not become more confident, it has stopped telling you the truth.
From the ZOKRI OKR Handbook, the methodology we install and maintain.
Reading about method is not the same as running it. We install this system and build the capability that stays.
Coach your teams to set better targets →