F27PEOPLE

Performance review drafting copilot

Helps managers write structured performance reviews by synthesizing the evidence already captured throughout the period: project outcomes, peer feedback, 1:1 notes, goals progress. Drafts a first version of each review section the manager edits to their voice and judgment. Catches common pitfalls (one-incident bias, recency effect, vague language, missing examples) and prompts the manager to add specifics. Never makes the rating decision — that stays entirely with the human. The pattern's value is reducing the slog of review-writing while improving review quality, particularly for managers with many direct reports who would otherwise write thin reviews under time pressure.

WHERE THIS FITS

BUSINESS SHAPES

B2B servicesProfessional servicesProduct company

VOLUME THRESHOLD

Below 30 reviews per cycle a month, the payback rarely earns the build. Patterns this shape reliably pay back at 200+.

REQUIREMENTS · 4 REQUIRED, 2 OPTIONAL

Requirements describe capabilities the pattern needs in your environment, not the vendors you must buy. Any system that fills a requirement satisfies it — that’s what makes the catalog portable across the long tail of SMB tooling.

employee_goals_and_evidence_corpus
REQUIREDREADcorpus
The evidence base for each employee's performance during the period.
DATA SHAPE
Per-employee: goals set at period start, project outcomes, 1:1 notes, peer feedback submitted, self-assessments.
COMMONLY FILLED BY
- performance management system with goals and check-ins
- 1:1 notes captured in a structured format
- peer feedback collected through a dedicated tool
- project archives showing outcomes
review_template_and_rubric
REQUIREDREADcorpus
The structure of the firm's reviews: sections, rating scales, calibration definitions, expected length.
DATA SHAPE
Per-role-or-level template with sections, rating definitions, behavioral examples per rating, calibration anchors.
COMMONLY FILLED BY
- performance review template in the HR system
- structured rubric co-authored with People Ops
- competency framework with behavioral indicators
manager_working_surface
REQUIREDREAD + WRITErequest
Where the manager writes the review. Pattern drafts inline; manager edits.
DATA SHAPE
Review document or form with sections; pattern fills draft text in each section and provides evidence references in side panel.
COMMONLY FILLED BY
- review interface in the HR system
- shared document with structured template
- purpose-built review-writing tool
bias_pattern_detector
REQUIREDREAD + WRITErequest
Watches for common pitfalls in the manager's draft: vague claims, evidence imbalance, language patterns associated with bias.
DATA SHAPE
Detected patterns with the specific text, the concern, suggested revisions or prompts for the manager.
COMMONLY FILLED BY
- bias-language module built for the engagement
- structured checklist of pitfalls the pattern checks against
- review-quality detector trained on prior reviews
peer_feedback_aggregator
RECOMMENDEDREADcorpus
Where peer feedback gets synthesized for the manager's view, with appropriate anonymization.
DATA SHAPE
Aggregated peer feedback with themes identified, anonymized at appropriate aggregation level.
IF MISSING
Reviews still incorporate peer feedback if available raw, but synthesis is manual. Recommend for any firm running 360-style reviews.
COMMONLY FILLED BY
- peer feedback collection in the HR system
- anonymized aggregator built for the engagement
- structured feedback library tied to each employee
calibration_destination
RECOMMENDEDWRITEbatch
Where draft reviews feed into the calibration process so leadership can see distributions and consistency across managers.
DATA SHAPE
Per-manager draft reviews with proposed ratings, evidence quality assessments, distribution statistics.
IF MISSING
Reviews still happen but calibration is harder. Recommend for any firm with formal calibration processes.
COMMONLY FILLED BY
- calibration view in the HR system
- structured prep deck for calibration meetings
- spreadsheet collated for HR business partners

RUNTIME FLOW · 8 STEPS

01
Manager initiates a review for an employee
manager_working_surface
02
Pull the evidence corpus for the period: goals, project outcomes, 1:1 notes, peer feedback, self-assessment
employee_goals_and_evidence_corpus
03
Synthesize peer feedback into themes with appropriate anonymization
peer_feedback_aggregator
DECISION Skip if peer_feedback_aggregator not filled.
04
Load the appropriate review template and rubric
review_template_and_rubric
05
Draft each section: accomplishments, strengths, growth areas, goals progress, suggested ratings (as draft only, not a recommendation)
06
Place drafts in the manager's working surface with evidence references for each claim
manager_working_surface
07
Run bias pattern detection on the manager's edits, flag issues inline with suggested revisions
bias_pattern_detector
08
On submission, contribute to calibration destination for cross-team consistency review
calibration_destination
DECISION Skip if calibration_destination not filled.

EMISSIONS · 3

Structured outputs this pattern produces. Other patterns and client systems can subscribe to them, which is how the catalog composes over time.

review_quality_signal
Per-review: length, evidence density, specificity score, bias pattern flags before/after manager edits.
CONSUMED BY
- People Ops quality reviews
- manager coaching priorities
- review process refinement
evidence_corpus_richness
Per-employee tracking of how much performance evidence was captured during the period, useful for spotting employees whose work is invisible to the system.
CONSUMED BY
- manager 1:1 prep
- employee development planning
- visibility for under-recognized contributors
calibration_input_signal
Distributions of draft ratings before calibration discussions, useful for prepping calibration meetings.
CONSUMED BY
- calibration meeting prep
- HR business partner workflows
- leadership consistency reviews