Feedback & Calibration Session Guide

Welcome

Calibration is a structured conversation that helps teams align expectations, reduce rating drift, support fair decisions, and surface development opportunities. This guide helps you run calibration sessions that are efficient, evidence-driven, and psychologically safe—so ratings reflect observable behaviors and everyone leaves clearer about next steps.

What success looks like

• Consistent, evidence-based ratings across similar roles and contexts.
• Clear development actions for each person discussed.
• Preserved trust: conversations feel constructive, not punitive.
• Reduced appeals and disputes because ratings are documented and defensible.

Who should attend and their roles

• Moderator/Facilitator: Keeps time, enforces agenda, and ensures psychological safety.
• Raters/Managers: Present cases for their direct reports with evidence.
• Calibration Panel (peers or senior leaders): Provide perspective and surface bias or inconsistency.
• Recorder: Captures evidence, final decisions, and action items.

Preparation (pre-work — required)

1. Collect evidence for each person: recent work samples, objective metrics, peer feedback, customer feedback, and examples of behavior.
2. Have managers complete a one-page summary for each person (see template below).
3. Distribute a short read-ahead that describes the calibration rubric and anchor examples.
4. Limit the number of people to discuss so each case gets adequate attention (aim for 6–10 cases per 90-minute session).

Suggested Agenda (90 minutes)

1. 0–10 min: Welcome, objectives, and ground rules (psychological safety, confidentiality, evidence-only discussion).
2. 10–15 min: Quick rubric refresher and anchor examples.
3. 15–75 min: Discuss cases (8–10 minutes per case): manager presents evidence (2–3 min), panel questions (3–4 min), calibration decision and development actions (1–2 min).
4. 75–85 min: Record disputes and unresolved cases; assign owners for follow-up.
5. 85–90 min: Close — confirm next steps, document changes, schedule appeals if needed.

Evidence-Collection Template (one-page manager summary)

Use this short template to keep presentations focused and comparable.

Anchoring Examples (rating anchors)

Provide short, behavior-focused anchors so raters use the same yardstick.

• Exceptional — Consistently exceeds expectations: delivers reliably under ambiguity, drives measurable results, coaches others; objective evidence and multiple stakeholders confirm impact.
• Solid Contributor — Meets role expectations: reliable delivery, some stretch impact, occasional leadership on tasks; evidence shows consistent performance with some areas to grow.
• Developing — Inconsistent delivery or needs support in key areas; evidence shows capability but gaps remain affecting outcomes.
• At-risk — Performance gaps are affecting team/customer outcomes; immediate development or performance plan recommended.

Dispute Resolution Process (fast, fair, documented)

1. If a panelist disagrees with a recommended rating, ask them to cite concrete evidence that supports a different view.
2. Allow the presenting manager two minutes to respond with additional context or evidence.
3. If disagreement remains, the panel records it as a "dispute" and assigns a short path: (a) seek missing evidence within 3 business days, (b) reconvene a 15-minute adjudication meeting with an agreed adjudicator, or (c) escalate to HR for final determination if policy is implicated.
4. Always document both positions and the rationale so follow-up conversations with the employee can be factual and constructive.

Action Items & Follow-up (what to record and by whom)

• Final calibrated rating and the rationale — Recorder
• Development actions, owner, and target dates — Manager
• Any agreed changes to role or compensation — HR liaison
• Disputes and next steps — Facilitator

Recommended Frequency & Cadence

Calibration timing depends on organization size and rhythm:

• Fast-moving teams or those with quarterly reviews: calibrate before each review cycle (quarterly).
• Smaller teams or stable roles: semi-annually may suffice.
• Ongoing informal calibration (monthly peer checks) can reduce heavy review sessions and rating drift.

Tips for Psychological Safety and Bias Reduction

• Start with positive examples and observable facts; avoid personality labels.
• Encourage questions that seek evidence, not opinions.
• Rotate panel membership regularly to avoid entrenched biases.
• Call out common biases (recency, halo, leniency) at the start of each session.

Common Pitfalls and How to Avoid Them

• Too many cases per meeting — prioritize quality over quantity.
• Long, unstructured presentations — require the one-page template.
• Decisions made without evidence — pause and request documentation.
• Punitive tone — explicitly frame calibration as development-first unless policy reasons require otherwise.

Small-team and Remote Adaptations

For small teams, combine calibration with coaching huddles and keep sessions under 60 minutes. For remote sessions, share read-aheads and templates in advance, use breakout rooms for parallel case prep, and capture decisions in a shared document with versioning.

Metrics to Track Over Time

• Rating distribution changes (to detect drift).
• Number and type of disputes per cycle.
• Percentage of development actions completed on time.
• Manager and employee perceived fairness (short anonymous pulse after cycle).

Next steps & templates

Start by running a single pilot session with a small group. Use this guide as the agenda. Capture lessons and update the one-page template. Consider making the templates interactive so managers can submit summaries ahead of time and the recorder can capture decisions directly into the system.