Formula
| Term | Meaning |
|---|---|
| Employees with the rating | Employees whose final rating after calibration in the cycle was that point on the scale. |
| Total employees rated | Everyone who received a final rating in the cycle. Leave out people not eligible, such as very recent joiners, and state the rule. |
Worked example
A manufacturing company in Aurangabad rated 200 employees on a 5-point scale for FY 2025-26. Final ratings after calibration were 16 employees at 5, 54 at 4, 96 at 3, 26 at 2 and 8 at 1.
- Total rated = 16 + 54 + 96 + 26 + 8 = 200
- Rating 5 = 16 / 200 x 100 = 8%; rating 4 = 54 / 200 x 100 = 27%
- Rating 3 = 96 / 200 x 100 = 48%; rating 2 = 26 / 200 x 100 = 13%; rating 1 = 8 / 200 x 100 = 4%
- Average rating = (5 x 16 + 4 x 54 + 3 x 96 + 2 x 26 + 1 x 8) / 200 = 644 / 200 = 3.22
Reading the shape of the distribution
Look at the shape, not only the average. A distribution bunched at 3 suggests managers avoid difficult conversations at both ends. One bunched at 4 and 5 suggests leniency, which makes top ratings meaningless and strains the increment budget. Compare the distribution with your own previous cycles and between departments of similar size. There is no correct shape to copy from elsewhere; what matters is that ratings reflect real performance and are applied consistently.
Forced distribution versus calibration
Some companies force ratings into fixed percentages for each band. That controls leniency, but it can push a strong team to label someone a low performer just to meet the curve. Calibration is the gentler alternative: managers bring proposed ratings to a meeting, compare evidence across teams and adjust outliers. Whichever you choose, write the rule into the performance policy and apply it the same way in every department.
- Distribution by department and manager.
- Distribution by gender and grade, to check fairness.
- Change in each person's rating from the previous cycle.
Pitfalls in rating data
Small teams cannot be expected to match a company-wide shape, since a team of four may genuinely have two strong performers. Compare large groups, and review small teams on evidence rather than percentages. Leave out people who joined too recently to be rated, and say so. Keep the scale fixed between cycles, because switching from a 5-point to a 4-point scale breaks every comparison with earlier years.
How to improve it
- Define each rating level with behavioural examples so managers read the scale the same way.
- Hold calibration meetings before ratings are shared with employees.
- Ask managers to give written evidence for top and bottom ratings.
- Compare distributions by department, manager and gender every cycle.
- Link ratings to documented goals agreed at the start of the cycle.
Tracking it in ZeniaHR
ZeniaHR's Performance module sets up review cycles with a rating scale and weighted goals for each employee, and the cycle summary shows reviews by status. For new joiners, probation reviews carry a manager rating of 1 to 5 from My Team. ZeniaHR has no self-appraisal or manager review forms, so collect final ratings after calibration in a spreadsheet and calculate the share at each point of the scale.
See it on your own data
A 30-minute demo on a video call. We set up your departments, shifts and leave rules and show attendance, leave and payroll running for your team. Free for your first 50 employees.
Book a free demoSee pricingFrequently asked questions
How do you calculate performance rating distribution?
Count the employees who received each rating in the cycle, divide each count by the total number rated and multiply by 100. For example, if 96 of 200 employees were rated 3, that band holds 48 percent. The percentages for all rating points should add up to 100.
What is a normal performance rating distribution?
There is no single normal distribution that every company should match. Ratings should reflect real performance and be applied consistently across teams. Compare your distribution with your own previous cycles and between departments, and look into any team whose ratings sit far above or below the rest without clear evidence.
Is forced ranking a good idea?
It controls rating inflation, but it can force a strong team to rate someone low only to fit the curve, which damages trust. Many companies prefer calibration meetings, where managers compare evidence and adjust outlier ratings without fixed quotas. Whichever approach you use, explain it to employees in advance.