Workers

Worker Reliability Scores

Understand how TempClock calculates worker reliability across 22 metrics, how to read scores, trends, and risk flags, and how AI-powered insights help you make smarter workforce decisions.

6 min read

What Are Reliability Scores?

AI-powered workforce intelligence at a glance

Worker Reliability Scores give you a single 0–100 rating for every worker, calculated automatically from 22 data points across 6 categories. Instead of relying on gut feeling or manually checking timesheets, you get an objective, data-driven view of who you can count on.

Automatic scoring — no manual input needed. TempClock analyses clock-in data, cancellations, compliance, incidents, and more
Scores update daily so you always have a current picture
Trend tracking shows whether each worker is improving, declining, or stable over the last 30 days
Risk flags alert you to workers who may need attention before problems escalate
AI-powered summaries give you plain-English workforce assessments at a glance
Score Range

0–100, where 100 is perfect reliability

Data Points

22 individual metrics across 6 weighted categories

Update Frequency

Recalculated daily based on the latest data

History

90 days of daily score snapshots for trend analysis

Note

Reliability Scores are available on the Enterprise plan. They appear in the admin dashboard under Reliability Scores in the sidebar, and tenant managers can also access a client-scoped version.

How Scoring Works

The calculation behind the number

Each worker's score is built from 6 weighted categories. Within each category, individual metrics are scored from 0 to 100 and combined using their own weights. The category scores are then blended using the category weights, plus a 5-point base, to produce the final composite score (capped at 100).

The formula is: composite = sum(category_score × category_weight) + 5, capped at 100.

A: Attendance & Punctuality

30% weight — the biggest factor

B: Shift Commitment

20% weight

C: Responsiveness

10% weight

D: Compliance & Documentation

15% weight

E: Quality & Conduct

15% weight

F: Engagement & Flexibility

10% weight

Tip

The 5-point base means even a worker with some struggles won't hit absolute zero. This prevents a single bad week from making a score look catastrophic.

The 6 Scoring Categories

22 metrics organised into weighted groups

Here's what each category measures and why it matters.

A. Attendance & Punctuality (30% of total)

The single biggest factor. Measures whether workers show up and show up on time.

On-Time Rate (30%) — clock-in within 15 minutes of shift start
Arrival Variance (25%) — average minutes early or late
Early Departure Rate (20%) — clocked out more than 15 minutes early
No-Show Rate (20%) — scheduled but never clocked in
Attendance Streak (5%) — bonus for consecutive shifts attended

B. Shift Commitment (20% of total)

Tracks whether workers follow through on accepted shifts.

Completion Rate (35%) — shifts completed vs assigned
Cancellation Rate (30%) — overall cancellations
Short Notice Cancellations (20%) — cancellations within 24 hours
Weekend/Holiday Reliability (15%) — reliability on premium shifts

C. Responsiveness (10% of total)

How quickly workers respond to shift offers and messages.

Confirmation Speed (60%) — time to confirm shift assignments
Message Acknowledgment (40%) — percentage of messages acknowledged

D. Compliance & Documentation (15% of total)

Measures how well workers stay on top of required paperwork and procedures.

Document Compliance (40%) — critical documents valid and up to date
GPS Geofence Compliance (25%) — clock-ins within the geofence
Break Attestation (20%) — qualifying shifts with break confirmed
Hours Attestation (15%) — shifts with hours confirmed

E. Quality & Conduct (15% of total)

Behavioural track record including incidents and disputes.

Incident-Free Score (40%) — weighted by severity (low=1, med=3, high=7)
Dispute Rate (25%) — shift queries as percentage of entries
Adjustment Frequency (20%) — manual adjustments as percentage
Face Match Confidence (15%) — average facial recognition score

F. Engagement & Flexibility (10% of total)

How engaged and adaptable the worker is beyond their basic duties.

Survey Response Rate (40%) — surveys completed vs triggered
Multi-Location Flexibility (30%) — distinct locations worked
Availability Coverage (30%) — declared availability consistency
Note

Not all metrics will apply to every deployment. For example, if you don't use GPS geofencing, that metric won't negatively affect scores — it's automatically excluded from the calculation.

Reading the Scores Dashboard

Navigate the Reliability Scores page

The Reliability Scores page is your central command for workforce intelligence. It combines summary statistics, filters, a sortable worker table, and expandable detail panels.

Check the summary cards

At the top of the page, four summary cards give you the big picture:

  • Average Score — The mean reliability score across all active workers
  • Improving — Workers whose score has risen more than 2 points in the last 30 days
  • Declining — Workers whose score has dropped more than 2 points
  • At Risk — Workers with active risk flags that need attention

Use the filter tabs

Filter the worker table by status:

  • All Workers — Everyone with a calculated score
  • Reliable (80+) — Your most dependable workforce
  • Attention (55–79) — Workers who could improve with support
  • At Risk (below 55) — Workers requiring immediate attention

Read the worker table

Each row shows:

  • Score badge — colour-coded circle (green 80+, amber 55–79, red below 55)
  • Trend arrow — green up arrow (improving), red down arrow (declining), or grey dash (stable)
  • 6 category mini-bars — quick visual of performance across all 6 categories
  • Risk status — badges for any active risk flags

Expand a worker for full detail

Click any worker row to expand the detail panel, which includes:

  • Radar chart — Visual comparison of all 6 categories
  • 90-day trend line — Historical score chart showing improvement or decline over time
  • Category breakdown — Each category with its individual metrics and scores
  • AI assessment — A plain-English summary of the worker's performance from Ada, TempClock's AI
Tip

Click the column headers to sort the table. Sorting by individual categories helps you find workers who are strong overall but weak in a specific area — for example, someone with a high attendance score but low compliance.

TempClock compares each worker's score over the last 30 days against the previous 30 days to calculate trends and automatically flags workers who may need attention.

Improving

Score increased by more than 2 points over 30 days. Shown as a green up arrow.

Declining

Score decreased by more than 2 points over 30 days. Shown as a red down arrow.

Stable

Score changed by 2 points or less. Shown as a grey dash.

Risk flags are automatically raised when specific thresholds are crossed:

Recent high-severity incident — A serious conduct or safety incident was logged in the last 30 days
On-time rate drop — The on-time rate has fallen by more than 10 points recently
Cancellation increase — The cancellation rate has increased by more than 15 points
Important

Risk flags appear as badges in the worker table and in the detail panel. They're designed to catch problems early — a worker whose on-time rate is dropping may be heading for a no-show if the pattern isn't addressed.

AI-Powered Insights

Ada's workforce intelligence

When the AI module is enabled, TempClock's AI assistant Ada provides two types of intelligence on the Reliability Scores page:

Workforce Assessment

A 2–3 sentence summary at the top of the page covering overall workforce health, trends, and areas of concern. Updated hourly.

Worker Assessment

A single-sentence insight for each worker when you expand their detail panel. Highlights the most important thing to know about that worker right now.

Note

AI insights are optional and controlled by a feature toggle. If AI is disabled, the Reliability Scores page works exactly the same — you just won't see the Ada assessment cards. Scores, trends, and risk flags are all calculated without AI.

Tip

Ada's assessments are generated from the same data you see in the dashboard — scores, trends, incidents, and compliance status. They're designed to save you time by surfacing the most important patterns in plain English.

Best Practices

Get the most from reliability scoring

Follow these recommendations to make reliability scores a powerful part of your workforce management.

Review declining workers weekly — Use the Declining filter to catch negative trends early. A conversation now can prevent a no-show next week.
Recognise top performers — Workers with scores above 90 are exceptional. Use the data to reward reliability and reduce turnover.
Don't use scores in isolation — A low score is a signal to investigate, not an automatic reason to dismiss. Check the category breakdown to understand why the score is low.
Keep compliance data current — Document compliance is a significant factor. Ensure worker documents are uploaded and verified so scores accurately reflect reality.
Check risk flags daily — The At Risk summary card on the dashboard tells you how many workers need attention. Make it part of your morning routine.
Use trend data for shift allocation — Assign critical or high-value shifts to workers with stable or improving scores above 80.
Note

Reliability Scores work best when you have at least 30 days of clock-in data for a worker. New workers will have limited data, so their scores may fluctuate more in the first few weeks.

Ready to try TempClock?

10-day free trial. All features included. No hidden fees, cancel anytime.