Module 5 — Calibration Across Managers — Owning the Distribution

60-Second Summary

Module 5 of the Advanced Manager-of-Managers program. Theme: From writing a level case to defending a distribution.
Calibration plan for next cycle — the real artefact you produce.
4-hour monthly intensive + biweekly coached practice on real work.
Reviewed by CHRO, VP/Director, sitting CEO, and OB faculty lenses.

At your old layer, you wrote level cases and submitted them upward. At this layer, you OWN the calibration room — the distribution, the bias surfaces, the manager-coaching that follows. This is one of the highest-leverage and highest-risk meetings in the year: it determines comp, promotions, perceived fairness, and (in adverse cases) future litigation. Most directors run it on instinct. This module installs the discipline.

What the evidence says

Aguinis on performance management: calibration without explicit anchoring produces 15–25% rating variance across managers for equivalent performance.
Castilla & Benard on meritocracy paradox: rating systems amplify bias unless calibrators actively correct.
Bock (Work Rules!): Google's calibration discipline (anchored examples, distribution targets, bias call-outs) is the single most-cited people-system innovation.

Pre-read (90 minutes)

Aguinis — Performance Management, calibration chapters — 40 min.
Your company's calibration guide, rubric, and last cycle's distribution — 30 min.
Anonymised examples of 3 reports at the same target rating from different managers on your team — 20 min.

Monthly intensive (4 hours)

Cohort flow with a senior practitioner coach

1
Distribution literacy (45 min)
Walk through last cycle's distribution. Where did each manager cluster? Where are the suspicious patterns?
2
Bias surfaces (60 min)
Coach walks through the standard bias patterns: recency, halo, similarity, gender, race, parent-status. Cohort identifies which they've actually seen in their own calibrations.
3
Calibration run-through (60 min)
Cohort runs a simulated calibration with planted bias. Coach surfaces what they missed in real time.
4
Manager coaching after (45 min)
Calibration is upstream of manager-coaching. Walk through how to coach a manager whose ratings were way off the distribution.
5
Wrap (30 min)
Each leader names one calibration improvement they'll install this cycle.

The artefact you produce

Calibration plan for next cycle

Pre-read template, anchored example library, distribution targets, bias-surface checklist, post-calibration manager-coaching plan. Reviewed with HRBP.

Tools at this layer

Layer	Examples (2026)	Use
Calibration tooling	Lattice Calibration, Workday Talent, internal calibration doc	Visible distributions, comparable cases
Bias surfacing	Pre-circulated bias checklist, blind name review for first pass, demographic distribution review	Surface bias systematically, not by hope
Anchored examples	Internal library of 'this is what this rating looks like at this level' anonymised cases	Anchoring is the single largest lever on cross-manager consistency

Copy-paste AI prompt

Here's an anonymised calibration scenario [3 reports at the same proposed rating from different managers, with their evidence]. Help me: (1) identify likely bias patterns in the evidence, (2) draft anchoring questions to ask each manager, (3) suggest how I'd coach a manager whose distribution is significantly more lenient than peers.

Between-session homework

Last cycle's distribution analysed; patterns and outliers documented.
Pre-circulated calibration pre-read template designed.
Bias-surface checklist drafted with HRBP.
Calibration plan submitted to coach.

Success signal

By end of this module, you can defend your org's rating distribution to your CEO and your HRBP with evidence, name and correct the specific biases each of your managers tends toward, and run a calibration room where the quiet voices get heard and the loud anchoring gets neutralised.

Reviewer notes

CHRO (20+ yrs)

Calibration is where I learn whether a director can hold a room of strong personalities to a system. The ones who can become VPs; the ones who let the loudest manager dominate stay where they are.

VP / Director (15+ yrs, 3+ scaled orgs)

The hardest moment in calibration is overruling a manager whose evidence is thin but conviction is high. If you can't do that, you don't own the distribution — the loudest manager does.

Sitting CEO

When I look at a director's calibration outputs, I'm not looking at the distribution. I'm looking at whether the same level of work got the same rating regardless of manager. That number tells me everything.

OB / HR Professor (25+ yrs)

The procedural-justice and bias-amplification literatures converge here: rating systems are only as fair as their calibration discipline. Anchored examples and explicit bias surfaces are evidence-based interventions, not nice-to-haves.

Written by Pawan Joshi.Sources cited inline.

First published 23 Jun 2026See site changelog →