Ask a sales manager what happens to a rep’s forecast the week before quota gets locked in, and you already know the answer: it drops. Not because the pipeline got worse overnight. Because the rep just worked out exactly how much cushion they need to comfortably clear the number they’re about to be judged on.
That instinct has a name, and it’s older than OKRs. Sandbagging: deliberately underselling your own capability so the bar gets set low enough to clear easily, then quietly beating it. Poker players do it with their bankroll. Golfers do it with their handicap. Goal-setting teams do it with key results, every quarter, whether anyone admits it or not.
We wrote recently about why we don’t tie OKR outcomes to performance reviews, and the argument was simple: link scores to comp, and ambition shrinks. This is the other half of that story: what sandbagging actually looks like inside a real OKR programme, why it survives even in teams full of honest, hard-working people, and what a fix that isn’t just ‘try harder’ looks like.
What sandbagging looks like in a goal-setting context
Sandbagging in an OKR context rarely looks like lying. Nobody stands up in planning and says ‘I could hit 15% growth, but I’m going to promise 8% so I look good.’ It’s quieter than that. A product manager who genuinely believes a new onboarding flow could lift activation by 15% talks themselves down to an 8% target ‘to be safe’, hits 9%, and gets a glowing review.
The 15% idea never gets tested, because nobody was ever asked to commit to it. The target was technically hit. The business never found out what was actually possible.
Why smart, honest people do it anyway
It’s tempting to treat sandbagging as a discipline problem: hire more ambitious people, coach harder, set a higher bar in the planning meeting. That misses what’s actually going on. Sandbagging isn’t a character flaw. It’s the predictable output of a specific incentive design, and it shows up in sales forecasting and procurement negotiations for the exact same reason it shows up in OKRs (see Investopedia’s breakdown of sandbagging in business): whenever the number you commit to also determines your pay or your rating, the rational move is to commit to a number you’re confident you’ll beat.
This is exactly the failure mode management by objectives ran into decades before OKRs existed. Tie the annual objective to the annual review, and ‘agree on ambitious goals together’ quietly turns into ‘negotiate a goal you can definitely hit’. OKRs were meant to fix that by shortening the cycle and normalising a 70% hit rate as a good outcome. Plenty of companies have quietly rebuilt the same trap by wiring OKR scores straight back into comp.
The three costs sandbagging quietly imposes
None of this shows up as a single dramatic failure. It shows up as three slow leaks.
- Planning gets built on numbers that are already discounted. Leadership plans headcount, budget, and roadmap around targets that were deliberately set below what the team believes is achievable, so the whole business plans smaller than it needs to.
- The ceiling quietly drops every quarter. Whatever number got rewarded this quarter becomes next quarter’s ‘ambitious’ target, because nobody wants to be the one who suddenly commits to something harder than what worked last time.
- The score itself stops meaning anything. Once every key result lands somewhere between 95% and 105%, the score can no longer tell you whether the team is calibrating well or just protecting itself. Good quarters and ordinary quarters become indistinguishable.
How to tell if your OKRs are already sandbagged
The tell isn’t a single missed target. It’s the shape of the distribution.
A few concrete signals worth checking before assuming everything’s fine:
- Almost every key result lands between 90% and 110%, quarter after quarter.
- Targets barely move even when the team, the product, or the market clearly changed.
- Nobody ever misses badly, and nobody ever blows a target out of the water either.
- In hindsight, every ‘stretch’ key result turns out to have been comfortably achievable.
If two or more of these are true across most of the team, the scores are telling you about risk tolerance, not performance.
A four-part fix for a sandbagging culture
None of this gets fixed by asking people to ‘be more ambitious’ in the next planning cycle. It gets fixed by changing what the target actually costs to miss.
- Audit the distribution before you audit anyone individually. Look at the spread of key result scores across the whole team over the last four quarters, not each score in isolation. A tight cluster is a system signal, not an individual one.
- Reward the honest miss when the effort was real. Someone who set a genuine stretch target, worked hard, and landed at 65% should come out of a review conversation looking better, not worse, than someone who set a safe target and cleared it at 101%. Say this out loud, more than once. It’s the same instinct that pushes people toward fake positive OKR updates instead of flagging a miss early. Take away the penalty for an honest miss, and both problems shrink together.
- Lock in the target before the comp conversation is anywhere near the table. The closer target-setting sits to a performance or pay decision, the more defensively people set it. Publish draft key results as early in the cycle as possible, while the only thing on the table is what’s actually true.
- Keep the review conversation separate from the check-in conversation. This is the same separation we’ve argued for before: weekly or fortnightly check-ins are about progress and blockers, not judgement, while performance reviews happen on their own cadence and look at the fuller picture. This is exactly the kind of discipline a proper StratOps cadence is built to enforce, rather than leaving it to whoever remembers to keep the two conversations apart that quarter.
A quick example
Score the raw percentage and Rep A wins easily. Look at what each person actually taught the business about what’s possible, and Rep B is the one worth promoting.
Where Tability fits
Spotting a sandbagging pattern requires seeing the distribution, not just this quarter’s number. Tability keeps a running history of every key result score across every team, so a hit rate that’s suspiciously tight around 100% for four quarters running is visible at a glance, rather than buried inside twelve separate check-ins nobody’s cross-referencing. It doesn’t solve the incentive problem for you (nothing does), but it does make the pattern impossible to miss once it’s there.
Keep your OKRs honest
If your team’s key results all seem to land suspiciously close to 100% every quarter, that’s worth a closer look before you call it a good quarter.
Sign up free or book 30 minutes with us and we’ll show you what a distribution actually reveals about your OKR programme.
.png)
.jpg)
