Star Rating Calculation Guide

How rep ratings work, how they're calculated, and how to override them

Overview

Every IC Rep receives a 0–5 star rating that reflects their real-world performance across scheduled visits, on-time completion, report compliance, and photo quality. The system calculates a score automatically from call report data and AI scans. Regional Managers and Admins can manually overridethe calculated score at any time — the override takes precedence and the system score becomes a visible suggestion.

4.5+ ★ — Excellent
3.5+ ★ — Strong
2.5+ ★ — Fair
1.0+ ★ — Needs Improvement
0.0+ ★ — Poor
Unrated — No visit data

How the Score is Calculated

The system-calculated score is a weighted average of five metrics, each converted to a 0–5 scale, minus a flag penalty:

On-Time Completion

25% weight

Percentage of assigned visits completed and submitted without late flags or human review triggers.

Compliance Score

20% weight

Average AI compliance score across all submitted call reports (planogram accuracy, correct answers, required fields).

Photo Quality

15% weight

Average AI photo quality score — checks for clarity, relevance, duplicates, and whether photos show the required before/after state.

Visit Completion Rate

15% weight

Ratio of completed/submitted reports to total assigned visits. Missed or abandoned visits lower this metric.

On-Site Check-In

10% weight

Percentage of assignments where the rep checked in on-site (check_in_time recorded). Validating physical presence is critical — not checking in directly damages the overall rating.

Flag Penalty
Deduction

Each flagged report (AI duplicates, irrelevant photos, human review required, or needs-correction status) deducts 0.25 stars from the calculated score, capped at a maximum of 1.5 stars.

FORMULA

calculatedScore =
  (onTimeScore × 0.25) +
  (complianceScore × 0.2) +
  (photoQualityScore × 0.15) +
  (completionScore × 0.15) − flagDeduction

Manual Override & Adjustment

Who can override?

  • Regional Managers — can rate any rep in their workgroup
  • Admins — can rate any rep in the system
  • Reviewers — can rate reps during quality review

How to override

  1. Open the rep's profile page (/rep-profile/:id)
  2. Find the "Your RM Rating" section in the Performance & Rating card
  3. Click the desired star (1–5) — the rating saves instantly to the rep's UserProfile
  4. An AuditLog entry is automatically created recording the old value, new value, who made the change, and when
  5. The override takes precedence immediately — the system-calculated score is shown as a reference below the override badge

When to override

  • Exceptional performance not captured by AI (e.g., short-notice coverage, going above and beyond)
  • Correcting an unfairly low AI score due to a known scanning error
  • Penalizing a rep for behavior not visible in visit data (e.g., unprofessional conduct, no-shows without formal flags)
  • Resetting a rating — to remove an override, contact an admin to clear the rm_rating field

Activity & Adjustment Log

Every rating change — whether a new rating, an adjustment, or a reset — is recorded in the AuditLog entity. Each entry captures:

Date & Time

When the action occurred

Action

e.g., 'Manual Rating Adjustment'

Old → New Value

Previous and new star rating

Details

Human-readable summary of the change

Actor

Who made the change (Admin, RM, or System)

Target User

Which rep the action was about

The log appears at the bottom of each rep's profile page and can be exported as CSV for compliance records.

Where the Data Comes From

CallReport

status, ai_compliance_score, ai_photo_score, human_review_required, ai_irrelevant_photos, ai_has_duplicates

Core visit metrics — compliance, photo quality, flags

ProjectAssignment

status, scheduled_date, check_in_time, check_out_time

Visit scheduling, on-time tracking, completion status

UserProfile

rm_rating, rm_rating_by, rm_rating_at

Manual override storage — the authoritative rating field

AuditLog

action, old_value, new_value, details, target_user_id

Complete history of every rating adjustment