Practice

Performance appraisal that people take seriously

This guide describes a review process, not a guarantee of fair ratings. Managers should assess evidence consistently and document the reasons for decisions.

Most appraisal systems fail for the same reason: they ask managers to remember a year in one afternoon. Fix the evidence problem and the rest of performance management becomes straightforward.

RigoHR team · Published 24 Jun 2026 · Edited 7 Sep 2026 · 11 min read

Ask employees what is wrong with their appraisal and you will hear three complaints: the goals were vague, the rating felt personal, and nothing happened afterwards. Each has a design fix, and none of them requires a longer form.

Quick answer An appraisal people trust combines measurable goals, role-based competencies, evidence collected throughout the cycle, calibrated review stages and a visible connection to development, promotion or reward decisions.

Start with goals that can be scored

A goal that cannot be measured becomes an opinion at review time. Every objective needs a measure, a target and a weight, and it should trace up to something the organisation is trying to achieve. For roles without natural numbers, define the standard instead: what does completed, on time and to quality look like for this role.

Cascade from organisational objectives so the link is visible to the employee
Set weights so priorities are explicit rather than implied
Agree targets at the start, and record any mid-year change with a reason
Keep the number of goals small — five that matter beat twelve that do not

Anchor to the job description

If the appraisal form and the job description describe different jobs, employees notice immediately. Assign the JD and its knowledge, skills and abilities formally, record acceptance, and build the competency section of the review from the same source. This also gives you a clean basis for identifying training needs later.

Collect evidence through the year

This is the part that changes appraisal quality most. Regular one-to-one check-ins, recorded feedback, appreciation and goal progress updates create a trail. At review time the manager summarises what is already documented instead of reconstructing a year from memory.

A simple rule Nothing in the annual review should be new information to the employee. If it is, the check-in cadence failed, not the employee.

Design the cycle stages deliberately

Self review first, so the employee frames their own year
Manager review with ratings against goals and competencies
Reviewer or skip-level input where the structure requires it
HR calibration before publication
A documented conversation, then publication

Calibrate before you publish

Different managers rate differently. Reviewing distribution across departments before results are released surfaces the manager who rates everyone highly and the one who rates nobody above average. Calibration is not forced ranking — it is a check that the same rating means roughly the same thing across the organisation.

Close the loop, or do not bother

An appraisal that does not affect anything trains people to treat it as paperwork. Outcomes should visibly reach at least one of: increment and pay revision, promotion or role change, and a learning plan built from the competency gaps the review identified.

Where AI helps, and where it does not

Writing is the slow part of a review cycle, and reading is the slow part for HR. RigoHR can draft review feedback from the ratings a manager has given, and summarise long review responses so a cycle can be read in an afternoon. What AI does not do is decide the rating. The manager's judgement, and the evidence behind it, stays where it belongs.

Running it in one system

Goals, KPIs, one-to-ones, feedback, job descriptions, review cycles and calibration all sit on the same employee record in RigoHR — the same record payroll and learning read. That is what lets a rating become an increment with arrears calculated correctly, or a competency gap become an assigned course with an exam and a certificate.