Performance Calibration: How to Run Fair Rating Sessions Across Teams

Performance calibration addresses a basic problem in performance management: different managers interpret rating scales differently. One manager’s “exceeds expectations” is another’s “meets expectations”. Without calibration, ratings — and the pay, bonus and promotion decisions linked to them — depend more on who your manager is than on how you performed. This guide explains what calibration is, why it matters, how to prepare and run a session, how to avoid bias, the role of rating distributions and how to communicate outcomes.

What is performance calibration?

Performance calibration is a process in which a group of managers, facilitated by HR or a senior leader, reviews proposed performance ratings for their employees, discusses the evidence behind them and adjusts ratings where necessary so that standards are consistent across teams. It typically happens before ratings are finalised and linked to pay or promotion decisions.

Calibration is part of the annual or semi-annual performance review cycle. It is most common in organisations where ratings drive bonus allocations — see our guide to variable pay and bonus schemes — or where promotion decisions require comparison across teams. It also helps identify high performers who deserve development opportunities and performance concerns that need support.

Why does calibration matter?

Calibration matters because managers vary in how strictly they rate. Some are lenient, some severe, some rate everyone in the middle. Without calibration, employees doing similar work at similar levels receive different ratings, and pay and promotions become unfair. Calibration also exposes bias and improves managers’ judgement over time.

Fairness perceptions strongly influence engagement. Employees who believe ratings depend on their manager’s habits rather than their performance lose trust in the whole system. Consistent ratings also protect the organisation: pay and promotion decisions based on calibrated, documented evidence are much easier to defend, including under pay transparency rules discussed in our pay equity and transparency guide.

How do you prepare for a calibration session?

Before the session, publish a clear rating guide with definitions and examples, ask managers to propose ratings with brief evidence for each employee, compile the data into a view by team and level, flag outliers and borderline cases, and share materials in advance so managers can review their peers’ proposals.

Evidence is the core. For each proposed rating, managers should summarise results against goals, key contributions, behaviours aligned with values and any relevant feedback — for example from 360 feedback or peers. Notes from regular one-on-one meetings make this much easier. Ratings without evidence should not be accepted into the session.

How do you run a calibration meeting?

Set ground rules, review the overall distribution, then discuss employees by level or role group rather than by team. Focus time on the highest and lowest ratings and borderline cases. Each manager presents evidence briefly; others ask questions and compare with similar employees. The facilitator keeps discussion on evidence and records decisions and reasons.

Useful ground rules include: discuss performance in the review period only, refer to evidence rather than personality, compare employees at the same level, and keep discussions confidential. The facilitator — often an HR business partner — should intervene when discussions drift into anecdotes, recency bias or comments about characteristics unrelated to performance. Keep the session time-boxed; long, exhausting sessions produce worse decisions late in the day.

Should calibration use a forced distribution?

Most organisations now avoid strict forced distributions, which require fixed percentages in each rating category regardless of actual performance. They can be demotivating, encourage internal competition and penalise strong teams. Instead, many use guideline distributions as a reference point that triggers discussion when a team departs significantly from it.

A guideline might suggest that roughly 10–20% of employees typically receive the top rating. If one team proposes 60%, the session asks why: is the team genuinely exceptional, or are standards being applied leniently? The answer may be either. The key difference from forced ranking is that evidence decides, not the quota.

How do you communicate calibrated ratings?

Managers should communicate final ratings in a performance conversation, explaining the evidence and how the rating compares with expectations. If a rating changed during calibration, the manager should own the final decision and explain it in terms of consistent standards, without blaming the process or other managers.

Avoid telling an employee “I wanted to give you a higher rating, but calibration lowered it”. This undermines trust in both the manager and the system. Instead, explain what the evidence shows, acknowledge strengths and agree development priorities. Train managers in advance for these conversations; many find them the hardest part of the cycle. Our article on giving effective feedback covers practical techniques.

How can calibration be improved over time?

Review each cycle: how much ratings changed, which managers’ proposals moved most, how long sessions took, how employees perceived fairness, and what equity analysis showed. Use the findings to refine rating definitions, train managers and simplify the process.

Some organisations have moved to simpler rating scales or removed ratings altogether, relying on continuous feedback and calibrated pay decisions instead. Whatever approach you choose, the principle remains: decisions that affect people’s pay and careers should be consistent, evidence-based and explainable. Performance management software can streamline data collection and analysis; see our comparison of performance management tools.

How do calibration outcomes link to pay and promotion?

In many organisations, calibrated ratings feed into merit increases, bonus allocations and promotion decisions. Making this link transparent — through a merit matrix or bonus multiplier tied to ratings — helps employees understand outcomes, but also raises the stakes of calibration and the need for rigour.

Keep the processes connected but distinct. Calibrate performance first, then apply pay guidelines that consider rating, position in range and budget. Mixing pay conversations into the calibration session tends to bias ratings towards available budget rather than actual performance. For promotions, calibration can identify candidates, but readiness for the next level should be assessed against criteria for that level, not just current performance. Our guides to base pay and pay structures explain how merit matrices work.

What role does HR play in calibration?

HR designs the process, prepares data, facilitates sessions, challenges unsupported ratings, monitors bias, documents decisions and runs equity analysis afterwards. HR does not decide ratings itself; managers own them, but HR ensures that the process is fair and consistent.

Effective facilitators are neutral, well prepared and confident enough to challenge senior managers. They know the rating guide thoroughly, have data on historical patterns and can spot when discussions move away from evidence. Training HR business partners in facilitation and bias awareness is a worthwhile investment.

What are common calibration mistakes?

Common mistakes include running sessions without clear rating definitions, accepting ratings without evidence, letting senior voices dominate, forcing distributions regardless of performance, spending too long on average performers, discussing pay instead of performance and failing to communicate outcomes well.

Another frequent issue is calibrating too late, after managers have already hinted at ratings to their teams. This creates difficult conversations when ratings change. Ask managers not to share provisional ratings before calibration is complete, and plan the timeline so that final conversations happen soon after.

How do you calibrate when teams are remote or distributed?

Remote and hybrid teams add a visibility challenge: managers may know office-based employees better than remote colleagues, and contributions from quieter or distributed team members can be underestimated. Calibration should explicitly check for proximity bias by comparing outcomes for remote, hybrid and office-based employees.

Encourage managers to bring concrete evidence of remote employees’ contributions — deliverables, customer feedback, peer input — rather than relying on impressions from meetings. Tools that record goals and feedback continuously help close the visibility gap. Our guide to managing hybrid teams covers wider practices that keep distributed contributions visible.

Can calibration work without numeric ratings?

Yes. Organisations that have dropped numeric ratings can still calibrate decisions such as pay increases, bonuses and promotions by comparing evidence across teams. The discussion focuses on outcomes and contributions rather than labels, but the principles — consistent standards, evidence and bias checks — remain the same.

Without ratings, clear criteria for pay and promotion decisions become even more important, and documentation of the reasons behind each decision is essential for transparency and fairness.

How should new managers prepare for their first calibration?

New managers should learn the rating guide, gather evidence throughout the period, practise summarising each person’s contributions in a few sentences and ask an experienced manager or HR partner to review their proposals beforehand. Preparation builds confidence and makes discussion more productive.

Encourage new managers to listen in the first sessions, ask questions about how others interpret the rating levels and treat the experience as development.

Over a few cycles, they develop a calibrated sense of what each rating looks like in practice, which improves their own day-to-day feedback as well.

Pairing each new manager with an experienced calibration buddy for their first cycle is a simple, effective way to accelerate this learning.

Consistency grows with practice.

Frequently Asked Questions