goal setting
Also called: setting goals
Agreeing on goals that are specific, hard but reachable, accepted by the person and followed by regular feedback on progress. Such goals reliably beat "do your best", provided the person has the skill and the goal doesn't push out what isn't measured.
Goal setting is one of the best-supported findings in work psychology: goals that are specific and hard but reachable lead to better performance than "do your best" or no goal at all. A clear target tells people what good looks like, focuses attention, and makes progress visible, which is motivating in itself.
Four conditions make a goal work. It is specific (a number, a date, a clear outcome). It is challenging, but the team believes it can be reached. It is accepted: people either propose it or at least understand why it matters, because an imposed goal gets compliance rather than commitment. And it comes with feedback: a short, regular look at progress, ideally for the team rather than as a public leaderboard.
Two limits matter as much as the conditions. For new, complex work, a performance target can make results worse, because people guess faster instead of learning; set a learning goal first (master five exception types, with review) and add the number later. And a goal focuses attention on what it names, so anything unnamed can suffer: pair each target with a guardrail, a second number that must not get worse, and read it before rewarding anyone.
Example
Klasna's Support lead needs a team goal for May to replace a closes-only bonus. The bonus month shows what a one-number goal did (six coordinators, four weeks each):
The working: 1,413 / 1,497 = 94.39%; 1,616 / 1,774 = 91.09%.
A goal that meets all four conditions, in four lines:
- Target: at least 94% of tickets closed without a reopen, every week of May.
- Guardrail: coordinators solve at least 44 hard tickets over the four weeks, as in March.
- Agreement: the team proposes the numbers in the Monday call; the lead explains why quality matters in the peak.
- Feedback: a team-level chart every Friday, no personal league table.
Dmytro, still new to refund disputes, gets a learning goal instead: handle the five common exception types by the end of May, with Olha reviewing his first ten.
Common mistakes
- "Do your best" instead of a target. It sounds kind but gives no way to tell whether you are on track, so people settle lower.
- Imposing the number. Without a reason or a voice in it, people comply with the goal instead of committing to it.
- A performance target for work people are still learning. Set a learning goal first; the number comes once the skill is there.
- One target, no guardrail. Whatever the goal doesn't name (quality, hard cases) quietly gets worse.
- Goals without feedback. A number nobody looks at until the end of the quarter can't guide anyone's week.
Learn it in the course
- Goals people accept · Specific, hard but reachable, agreed goals with progress feedback, and guardrails against what one metric does to behavior.