Short answer
A rating scale works when each point has an observable meaning against the employee's current level expectations. Keep the scale stable across the cycle, require evidence for each rating, and use calibration to resolve differences in interpretation.
Define what the rating scale does
The rating scale summarizes how evidence compares with published expectations. It does not replace the Function, and it should not measure personality, effort, popularity, or future potential.
Peasy HR uses the term rating scale. It does not use the word rubric for this control because the assessable standard is the competency framework and its Functions.
Write observable anchors
Define the center point first. Meets expectations should mean the person consistently demonstrated the behaviors required at their current level during the review period.
Define lower and higher points by the evidence, consistency, and scope shown. Do not build the scale from vague labels such as low, medium, and high performance.
- Below expectations: important current-level behaviors were not demonstrated consistently
- Meets expectations: evidence supports the expected current-level work
- Above expectations: repeated evidence shows broader or more complex work than the current level requires
A usable five-point structure
A five-point scale can separate substantial gaps, partial evidence, expected performance, sustained evidence beyond the level, and rare evidence far beyond the level. Every point still needs the matching Function expectation and review evidence.
Do not treat the numbers as arithmetic. Averaging unsupported competency ratings can hide a critical gap or inflate a review with work that was easier to observe.
Anchor the midpoint
3: Good performance.
3: Meets expectations. The evidence consistently matches the published behaviors for the person's current level.
Apply the scale consistently
Train managers with the same sample evidence, then compare how they rate it. Differences reveal unclear anchors before those differences affect employees.
Use the same scale across Functions where possible, while each Function supplies the role-specific expectations. During calibration, discuss evidence against those expectations rather than comparing raw numbers alone.
Keep the scale stable during the cycle
Publish definitions before assessments open and pin them with the Function version. A mid-cycle wording change makes earlier ratings hard to compare.
After sign-off, review disputed ratings and unclear anchors. Update the next version, communicate the change, and give managers a concrete example before the next cycle begins.