Calibration Meeting Facilitation Playbook

This playbook is typically run by an HR Business Partner, Talent Partner, or another neutral facilitator who is not directly advocating for one employee or team. The facilitator keeps the discussion on evidence, applies the same rating criteria across managers, and captures decisions and follow-ups. A senior HR leader or department head may co-facilitate when the group needs an escalation path for disputed ratings.

It fits annual review cycles, mid-year check-ins, promotion calibration, and any session where multiple managers need to align on ratings before communication goes out. It is especially useful when teams have different standards, when managers are new to performance review, or when ratings affect compensation or promotion decisions. It is not meant for one-on-one manager feedback conversations.

Most organizations use this playbook once per formal review cycle, with additional sessions for mid-cycle promotions or exception handling. If the company has multiple business units, it can be run separately by function, level, or region to keep the discussion manageable. The cadence should match the point at which ratings become final and before any employee communication is sent.

Managers should bring proposed ratings, supporting examples, and any relevant goal or competency evidence. The facilitator should also prepare the rating scale, calibration criteria, roster of employees, and a list of known edge cases such as new hires, leaves of absence, or role changes. Without that prework, the meeting tends to drift into opinion rather than evidence-based comparison.

It creates a repeatable execution plan for comparing ratings against shared standards instead of manager-by-manager preferences. The facilitator can use the same prompts, confirm gate, and decision log for each employee or team, which helps surface rating inflation, deflation, and inconsistent interpretation of performance levels. That makes the final outcome easier to defend and explain.

Yes. You can adapt the input_schema and discussion prompts for sales, engineering, operations, or corporate functions, and you can calibrate separately by job level, location, or job family. Many teams also customize the evidence fields to reflect role-specific outputs such as quota attainment, project delivery, customer outcomes, or leadership behaviors. The core flow stays the same even when the criteria change.

The biggest mistake is letting the session become a debate about personalities instead of a review of documented evidence. Another common issue is starting without a clear rating scale or without pre-read materials, which slows the meeting and creates inconsistent decisions. Teams also run into trouble when they skip a decision log, because unresolved items get lost after the meeting ends.

Ad-hoc discussions are faster to start, but they often produce uneven standards and undocumented decisions. This playbook gives the meeting a consistent structure, defined steps, and a clear output so managers know what to bring and what gets decided. If your goal is repeatable, defensible calibration, a structured playbook is much easier to scale than informal discussion.