Common failure modes

• Recency beats results: last month’s high‑visibility task overshadows 9 months of compounding impact.
• Proximity beats fairness: co‑located contributors get more airtime than remote peers.
• Vague praise: “great teammate” with no data loses to “increased conversion by 3.2%.”
• Glue work gets erased: documentation, mentoring, and process improvement don’t appear as commits.

What great looks like

• Evidence portfolios: concise, business‑tied accomplishments for every report.
• Bias checks on every rating discussion.
• Clear impact narratives: Context → Challenge → Actions → Results → Ripple effects.
• Consistency across teams: same bar, same language, same artifacts.

Inputs to bring

• Role expectations per level and examples of “meets/exceeds.”
• OKR contributions and measurable outcomes.
• Cross‑functional feedback (Product, Design, Support, Sales).
• Evidence of glue work: docs, mentoring, incident leadership.

Rules of the room

• Calibrate to role impact, not personality or popularity.
• One narrative per person, time‑boxed (e.g., 90 seconds + Q&A).
• Challenge with data, not anecdotes.
• Document rationale and next steps (development plans, stretch projects).

Pre‑meeting checks

• Recency bias: Do early‑year wins get same weight?
• Proximity bias: Are remote/hybrid folks represented with equal evidence?
• Halo/Horns: Are you over‑weighting a single standout success or mistake?
• Homophily: Are you favoring similar backgrounds or communication styles?

In‑room bias blockers (scripts)

• “To avoid recency bias, here’s the year‑long trendline and business outcomes.”
• “Let’s anchor on level expectations. Here’s evidence against each criterion.”
• “Visibility ≠ value. Here are cross‑functional outcomes others didn’t see.”
• “Let’s separate style from impact. The data shows X moved Y metric.”

6 rules for advocacy

• Lead with outcomes, then show the work.
• Compare to level criteria, not peers by name.
• One story per rating: avoid laundry lists.
• Invite challenge: “What risks or gaps do you see?”
• Be ready with artifacts (dashboards, customer quotes).
• Timebox and land the ask: “I’m proposing Strong‑Meets.”

Handling pushback (scripts)

• “We didn’t see this work.” → “Visibility was low; here are cross‑team outcomes and artifacts.”
• “One incident went badly.” → “Agreed, and across 10 incidents they led 7 with fast recovery; trendline is strong.”
• “They’re quiet in meetings.” → “Style aside, the mentoring program they launched reduced onboarding time by 30%.”
• “Let’s keep the rating flat.” → “Compared to the rubric, their scope and outcomes are at the next level; here’s the evidence.”

New hires and leave

• Calibrate based on time‑weighted expectations; don’t penalize for ramp or approved leave.
• Use trajectory evidence: speed of learning, scope growth.

Glue work

Mentoring, documentation, and process leadership often lack dashboards—bring artifacts and outcomes (e.g., faster onboarding, fewer incidents).

Communicate with care

• Explain the rating with evidence and rubric language—no mystery.
• For growth ratings, co‑create a concrete plan (scope, skills, milestones).
• For disappointments, validate feelings, then redirect to actionable next steps.

Retention risks

• For top performers, discuss growth runway (scope expansion, promotion timing).
• Be proactive: if market pull is likely, prepare a data‑backed case for adjustments.

Core calibration drills

• Performance Calibration Meeting — practice advocacy, bias checks, and handling objections.
• Recognizing “Glue Work” — learn to make invisible impact visible.
• Presenting Your Team’s Work — ship concise, executive narratives.

Surrounding conversations

• Performance Review for an Underperformer — set clear expectations with compassion.
• Setting Team Goals (OKRs) — align outcomes early to support fair ratings.
• Your Own Performance Review — model the behaviors you expect from your team.