Performance Review Season Checklist

Review season has a lot of moving parts beyond the writing: calibration meetings, a full calendar of scheduling, ratings that need to hold up under scrutiny, and delivery conversations that can go sideways if you're underprepared. This is the full-cycle checklist, from the weeks before the cycle opens to the follow-up after every conversation is done.

The Checklist

Work through it in order. Tick things off as you go and your progress is saved on this device, so you can come back to it over the days before the conversation rather than doing it all in one sitting.

1. Confirm the cycle timeline before it starts

   Review season runs smoothly when everyone, including you, knows what happens when. It runs badly when the calendar only exists in HR's head.

   • Get the full cycle timeline in writing: self assessment deadline, manager review deadline, calibration dates, and delivery window.
   • Block your own calendar against every one of those dates now, not the week each one arrives.
   • Confirm which parts are mandatory for your level (calibration attendance, peer feedback requests, written sign-off) so nothing gets missed by assumption.

2. Collect inputs early: self assessments, peer feedback, goals

   Everything downstream, your ratings, your calibration case, your delivery conversations, is only as good as the inputs you gathered.

   • Send self assessment requests as early as your process allows, with a real deadline, not a soft one.
   • Request peer or stakeholder feedback for anyone whose work crosses teams, since that's exactly the input you're least likely to have firsthand.
   • Review the goals set at the start of the period before you rate anyone against them. Rating against goals nobody re-read is where drift creeps in.

3. Draft ratings against evidence, not impression

   This is the point where you translate everything you've gathered into an actual number and narrative per person, done with enough rigor to survive calibration.

   • Rate each person against specific evidence: logged achievements, concrete outcomes, feedback you actually received, not a general sense of the year.
   • Write two or three sentences of reasoning behind each rating before the calibration meeting, so you're not constructing the logic live under pressure.
   • Check your draft distribution across the team. A distribution that's suspiciously flat, with everyone bunched in the middle, is usually a sign some of the harder calls haven't been made yet.
   • If you've been keeping an achievement log for each report throughout the cycle, tools like Perform Review's AI Performance Reviews can turn that logged evidence straight into a first draft of the review.

4. Prepare specifically for calibration

   Calibration is where your ratings get tested against other managers' standards. Walking in prepared is the difference between holding a defensible line and getting steamrolled or, just as bad, caving on something you were right about.

   • Bring the evidence and reasoning for every rating, not just the numbers, so you can answer "why" on the spot.
   • Anticipate which ratings are most likely to get challenged, usually your highest and lowest, and know your reasoning cold for those specifically.
   • Go in willing to move a rating when the evidence genuinely supports it, and able to tell the difference between that and just caving to whoever argues loudest.

5. Schedule delivery conversations as their own dedicated meeting

   A review conversation squeezed into a regular 1:1 or bundled with a comp discussion rarely gets the focus it needs.

   • Book a dedicated slot, roughly 45 minutes, specifically for the review conversation.
   • Schedule any compensation discussion separately, even if it's only a few days later, so the feedback isn't heard through the lens of a number.
   • Give the employee enough lead time to read their self assessment and prepare questions, not a same-day invite.

6. Deliver each review with the rating, the reasoning, and a path forward

   However the rating lands, the conversation is what the person actually remembers, far more than the written document.

   • State the rating clearly early in the conversation rather than building up to it, since guessing where it's headed is worse for most people than just hearing it.
   • Back it with two or three concrete examples per major point, not general adjectives.
   • Leave every conversation with at least one clear, specific goal or focus area for the next period, agreed out loud, not just written down after the fact.

7. Document and file everything the same week

   Notes and reasoning fade fast once the season is over, and thin documentation is exactly what makes a rating hard to defend if it's ever challenged later.

   • File the final rating, written reasoning, and any agreed goals in your system of record within days of the conversation, not weeks.
   • Send each person a short written summary of what was discussed and agreed, so there's no gap between the conversation and the record.
   • Note anything you personally want to track differently next cycle, like a report whose visibility gaps made rating harder than it should have been.

8. Follow through in the weeks after

   The cycle isn't actually finished when the last conversation happens. What you do in the following weeks is what tells people whether any of it meant something.

   • Check in on the goals or focus areas you agreed to before the next official check-in point, not only when the next cycle opens.
   • Revisit anyone whose rating was borderline or contested sooner rather than later, so it doesn't sit unresolved for months.
   • Look back at your own process: which parts of this cycle actually took too long, and fix that specific step before the next one starts.

That is the whole checklist

Everything above is general by design, so it holds for most people in most companies. The specifics of your own case are the part it cannot cover.