Why photo review matters
Photo Review is the dispatcher's single most important daily workflow. Every time a driver starts or ends a shift, they capture a set of inspection photos (typically six: front, rear, left side, right side, dashboard, and mileage). These photos create a visual record of the vehicle's condition. Your job is to review these photos, compare them against previous shifts, and issue a verdict. This process is how Fleet by Elevera catches new damage early, attributes responsibility accurately, and protects your organization from disputed claims.
Clear the Photo Review queue every day, ideally first thing in the morning. Unreviewed shifts create gaps in the chain of custody — if damage appears two shifts later and the intermediate shift was never reviewed, it becomes much harder to determine when and how the damage occurred.
The photo review queue
Navigate to Photo Review in the sidebar. The pending count badge on the sidebar item shows how many shifts are waiting for your review. The page is divided into two main sections:
Pending tab
The pending tab lists every shift that has inspection photos but has not yet been reviewed. Each entry shows:
- Driver name and profile picture
- Vehicle plate number, make, model, and year
- Shift start time and end time
- Status — the shift's completion status
- Photo count — number of inspection photos captured
Shifts are sorted by start time (newest first), so the most recent completions appear at the top. You can filter the pending list by vehicle or driver using the dropdown filters.
Review history tab
The history tab shows all shifts you (or other dispatchers) have already reviewed. Each entry displays:
- The shift's driver and vehicle information
- Verdict — PASS, FAIL, or NEEDS_INSPECTION
- Comment — Optional notes left by the reviewer
- Reviewer name and review timestamp
- Selected photo IDs — which photos were specifically flagged during review
This history is your audit trail. If a question arises about when damage was first noticed, you can trace it back through the review history.
Reviewing a shift — step by step
- From the pending tab, click on a shift to open the Photo Review Comparison view.
- The comparison view displays the current shift's inspection photos alongside the previous shift's photos for the same vehicle. This side-by-side layout makes it easy to spot new scratches, dents, or other changes.
- Examine each photo pair carefully. Look for differences between the two sets — new marks, missing parts, changed dashboard readings, or discrepancies in the mileage photo.
- Check the AI Analysis panel (if available) for an automated assessment. The AI examines each photo and provides a verdict with a confidence score.
- Based on your assessment, select a verdict: PASS, FAIL, or NEEDS_INSPECTION. You can optionally select specific photos that you want to flag and add a comment explaining your reasoning.
- Submit your review. The shift is removed from the pending queue and appears in the review history.
Side-by-side photo comparison
The comparison view is the heart of the review process. It shows two columns:
- Current shift — Photos from the shift you are reviewing, captured at the start or end of the driver's shift
- Previous shift — The most recent prior shift for the same vehicle that also has inspection photos
Each photo displays its type (front, rear, left, right, dashboard, mileage) and the timestamp when it was captured. Compare corresponding photo types between the two columns — for example, the current "left side" photo against the previous "left side" photo.
Pay special attention to the corners, wheel arches, bumpers, and mirror areas in the exterior photos. These are the most common locations for new damage. Also compare the mileage photo between shifts to verify it is consistent with the vehicle's expected daily usage.
If no previous shift with photos exists for the vehicle (e.g., it is the first shift ever recorded), the previous column will be empty. In this case, review the current photos on their own merit and look for any pre-existing damage that should be documented.
AI-assisted vehicle condition assessment
When AI photo analysis is enabled, Fleet's automated system examines the inspection photos and produces:
- Verdict — The AI's recommendation: typically PASS (no issues detected) or FAIL (potential damage found)
- Confidence score — A percentage (0–100%) indicating how certain the AI is about its verdict. Higher confidence means the AI found strong evidence supporting its conclusion.
- Reasoning — A text explanation of why the AI reached its verdict, describing what it observed in the photos
- Flagged photos — Specific photos the AI identified as containing potential issues, highlighted in the comparison view
- Tier — Either
SCREEN(a quick initial pass) orDETAIL(a more thorough analysis). When both are available, the detail tier takes priority.
The AI analysis is an assistant, not a replacement for your judgment. Always review the photos yourself, especially when the AI confidence is below 80% or when the AI flags specific photos. The AI may miss subtle damage, and it may occasionally flag reflections or shadows as potential issues.
Auto-reviewed shifts
In some configurations, shifts where the AI gives a high-confidence PASS verdict may be automatically marked as reviewed. These will appear in your review history with an auto_reviewed flag. You should periodically spot-check auto-reviewed shifts to ensure the AI is performing accurately.
The verdict system
After examining the photos and AI analysis, you must issue one of three verdicts:
| Verdict | When to use it |
|---|---|
| PASS | The vehicle appears in the same condition as the previous shift. No new damage, no concerns. |
| FAIL | You have identified new damage, missing equipment, or a significant discrepancy. This should trigger an incident investigation. |
| NEEDS_INSPECTION | Something looks suspicious but you cannot confirm damage from the photos alone. The vehicle should be physically inspected before its next shift. |
When issuing a FAIL or NEEDS_INSPECTION verdict:
- Select the specific photos that show the issue. This creates a clear record of which photos prompted your concern.
- Add a comment explaining what you observed. Be specific: "New scratch on left rear quarter panel, approximately 15 cm, not present in previous shift photos" is far more useful than "damage found."
A FAIL verdict does not automatically create an incident. After marking a shift as FAIL, navigate to the Incidents page and create a new incident manually, linking it to the vehicle and driver. You can reference the flagged photos in the incident report.
Comparing with previous shift photos
The power of the review system comes from the chain of comparisons. Each review compares against the most recent previous shift that has photos. Over time, this creates a continuous visual record:
- Shift A photos (baseline) → reviewed as PASS
- Shift B photos compared against Shift A → reviewed as PASS
- Shift C photos compared against Shift B → reviewed as FAIL (new scratch)
This chain makes it clear that the scratch appeared during Shift C, which determines which driver was responsible for the vehicle at the time the damage occurred.
If you break the chain by skipping reviews, you lose this ability. That is why clearing the queue daily is so critical.
Tips for efficient reviewing
- Review in chronological order when catching up on a backlog. Start with the oldest pending shifts and work forward, so each comparison is meaningful.
- Use the AI confidence as a triage tool. Shifts with high-confidence PASS from the AI are likely fine and can be reviewed quickly. Low-confidence verdicts deserve more careful examination.
- Keep comments concise but precise. Future you (or another dispatcher) will thank you when investigating an incident weeks later.
- Flag specific photos rather than leaving a blanket FAIL. This helps when the incident needs to be reported to insurance.
Why thorough photo review matters
- Photo review is the single most important daily task for a dispatcher. A thorough review catches damage within hours of it happening, while evidence is fresh and the responsible driver is identifiable.
- Bulk-approving photos without looking at them undermines the entire inspection system. It takes 30-60 seconds per shift to review properly — that investment prevents costly disputes later.
- When AI flags a photo, it's doing the first pass for you. Your job is the judgment call: is this real damage, a shadow, or a pre-existing mark?
Recommendations
- Use keyboard shortcuts (n/o/d) to speed through the queue without sacrificing thoroughness. With practice, you can review a shift in under 30 seconds.
- Process the review queue before lunch every day. A backlog of unreviewed shifts means vehicles may be going out with undetected damage.
- When you tag a photo as Suspicious, add a brief note explaining what caught your eye. This helps the owner (or another dispatcher reviewing your work) understand your reasoning.
- Compare today's photos with yesterday's using the side-by-side view. New damage that appeared overnight usually means something happened during the previous shift or at the depot.