Most performance problems are expectation problems. People rarely fail at a job they clearly understood.
Setting expectations means being explicit about what success looks like in a role: the outcomes, the standard, the timeframe and how it will be judged.
It happens at the start, not at review time.
Because most 'performance problems' turn out to be someone doing exactly what they thought was wanted. Nobody told them what good looked like, so they invented a definition — usually a reasonable one that differed from yours.
And because unclear expectations are unfair. It is difficult to succeed at a job nobody described, and being judged against an unstated standard is corrosive.
Unclear: 'You're running marketing. Get us more customers.'
Three months later the founder is frustrated: too much time on brand, not enough on outbound. The marketer is frustrated too: they built a foundation, as any marketer would.
Neither is wrong. Nobody said which.
Clear: 'You own marketing. In the next quarter, success is 40 qualified leads a month at a CAC under $300, from at most two channels. Brand work is not a priority this quarter — we can revisit in Q3. We will review the numbers every Monday.'
Now the marketer can succeed, and can push back before starting if they think it is the wrong goal — which is far more useful than discovering the disagreement in month three.
Note the last part: naming what is not a priority is what makes the rest unambiguous.
First-time founders often assume competent people will work out what is wanted. Competent people will work out something, and it will be shaped by their previous job rather than your current priorities.
The second mistake: only stating expectations when they have already been missed. That is a criticism, not an expectation.
The third: not saying what is not a priority. Without exclusions, everything looks equally important and people spread themselves across all of it.
'40 qualified leads a month at CAC under $300' rather than 'more customers'.
By when. A goal with no date cannot be missed or met.
This is the part that makes the rest clear, and it is almost always omitted.
What is looked at, how often. No surprises at the end.
'What does success look like to you in this quarter?' The gap between their answer and yours is the whole risk.
Two lines in a shared document. Memory diverges, especially about standards.
At the start of every role, every quarter, and every delegated project.
Do not over-specify how the work is done. Expectations are about outcomes and standards; the method is theirs, or you are back to task-assignment.
Apply this to your own startup in My Full Journey (free account).