If you've evaluated fleet management software in the last five years, every vendor has told you their geofencing is "industry-leading." Most fleets buy it, configure two zones, get 80 alerts a day, and quietly stop checking the dashboard.
That's not because geofencing is broken. It's because it's widely misunderstood. Done well, geofencing is the highest-leverage telematics feature you'll deploy — it cuts detention charges, surfaces real-time exceptions, and replaces a thousand status calls with one alert. Done badly, it's a notification firehose.
This piece is the practical, no-jargon guide we share with every customer evaluating ViaLoop Fleet.
What is geofencing?
A geofence is a virtual boundary drawn on a map — a polygon, a circle, or a route corridor — paired with logic that triggers when a vehicle (or any GPS-tagged asset) crosses it.
That's the entire concept. Everything interesting about geofencing is in the logic: what counts as a crossing, what events fire, how you handle false positives, and how the system tells you something useful.
The simplest example: draw a circle around your warehouse. The geofence fires an "arrived" event when a vehicle enters and a "departed" event when it leaves. You log both timestamps. Now you have data on yard dwell time without anyone manually noting it down.
How geofencing actually works
Three layers. They're worth understanding because the difference between a useful geofence and an annoying one usually lives at the third layer.
1. The boundary
Modern fleet platforms support three shapes:
- Circular zones — fastest to draw, defined by a centre and a radius. Good for points of interest like fuel stations or customer addresses.
- Polygonal zones — drawn freehand on the map. Necessary for shapes that aren't round: a port, a warehouse complex, a highway corridor, a city limit.
- Route corridors — a tube around a planned route. Triggers when a vehicle deviates by more than X metres from the approved path.
2. The trigger
The platform compares each incoming GPS ping against your zones. When a position transitions from outside to inside (or vice versa), it fires an event: entered, exited, dwelling (in-zone for longer than a threshold), or deviated (off-route).
Sounds simple. The hard part is the quality of that detection. GPS jitter, momentary signal loss, and overlapping zones can produce phantom enter/exit events. A good platform debounces these — typically requires two consecutive pings inside the zone before firing "entered."
3. The logic on top
This is where most platforms stop being useful. A "vehicle entered yard" event is data, not a signal. To turn it into something operations can act on, you need rules:
- Combine multiple events: vehicle entered yard and waited 45 minutes and hasn't exited → potential loading delay.
- Apply context: vehicle exited port zone after 6pm → no detention charge yet, ignore.
- Route the alert: this kind of event goes to dispatch; that kind escalates to the customer ops manager.
The platforms that win on geofencing are the ones that treat layer 3 as the product. The platforms that fail treat layer 1 as the product and let customers figure out layer 3 themselves.
Where geofencing actually pays off — four real patterns
After deploying geofencing across consumer drivers, last-mile fleets, cold-chain operators, and port logistics, here are the patterns where it consistently pays for itself.
Pattern 1: Detention and demurrage
The single highest-ROI use of geofencing is on container ports, customer yards, and distribution centres where you're billed for time over a free allowance. Geofence the customer's yard, log entry/exit, generate a detention report at the end of the month. You stop relying on customer-supplied timestamps that always favour the customer.
For a UAE-based container operator we work with, geofence-based detention reporting alone surfaced ~38% in disputed charges they were previously absorbing.
Implementation tip: Don't trust the GPS at the boundary edge. Geofence inside the customer's yard by 30m, so you're sure the vehicle is on premises before the dwell timer starts. Saves arguments.
Pattern 2: Customer milestone alerts
For B2B fleets, customers don't want to log into your dashboard. They want a Slack/Teams/SMS notification when their shipment hits each milestone — left port, cleared customs, arrived yard, departed for delivery. Each milestone is a geofence. Each crossing is a webhook to the customer's system.
This is the "replaces a thousand status calls" pattern. We've seen operations teams cut their inbound "where is my container" calls by 70%+ after wiring milestone alerts to customers directly.
Pattern 3: Route compliance and fuel theft
Route corridors flag deviation in real time. Useful for two distinct problems:
- Fuel theft — the most common pattern is a stop at a non-approved fuel station for 8–12 minutes. The route corridor catches this within 60 seconds.
- Personal detours — drivers occasionally re-route to a familiar refuel station, a stop at home, etc. Adds 8–15 km per trip; visible on the corridor.
Don't use route corridors as a punitive tool. Use them as a reporting tool. Drivers behave better when they know the data exists; they don't need to be flagged every single time.
Pattern 4: Geofence-based safety zones
For consumer customers and small fleet operators, "safe zones" (home, workplace, school) trigger different behaviours. Auto-arm when the car leaves the home zone after 11pm. Auto-disarm during weekday morning commute. Send a parent an alert when the car leaves a 2-km radius around school during school hours.
Across our consumer ViaLoop dataset, this is the highest-satisfaction use case for parents tracking teenage drivers. The 11pm Saturday alert is what they actually care about — not the Tuesday morning commute.
The pitfalls — what makes geofencing become noise
Most fleets don't fail at geofencing because the technology is bad. They fail because of one of these mistakes:
1. Too many small zones
We've seen fleets with 600+ active geofences. The interface becomes unmanageable. Operations stop interacting with the alert feed. Start with the 10 highest-value zones and expand based on what produces actionable signal.
2. Treating every crossing as an alert
A crossing is an event. An alert is a thing that requires human attention. Most events should not be alerts. They should accumulate quietly into a daily/weekly report. Use alerts only for exceptions: dwelling beyond threshold, deviation from corridor, entry into an unauthorised zone.
3. Ignoring valet, parking, and edge cases
Cars get valet-parked outside a geofenced home zone. Trucks idle outside a yard while waiting for a gate. These produce false-positive alerts that train operations to ignore the system. Build in "valet mode" (manual mute) and "arrival queue" awareness (don't fire detention until the gate-in event).
4. No escalation logic
A vehicle dwelling 30 minutes past expected departure is a soft signal. Two hours is a hard signal. Most platforms don't differentiate. Good platforms escalate: notify driver app first, then dispatch at +30 min, then ops manager at +90 min.
How to set up geofencing well — a checklist
If you're configuring geofencing on any platform, work through these in order:
- Pick your top 10 zones first. Customer yards with detention exposure, your own facilities, key transit points (port exits, customs zones).
- Define the dwell thresholds. What counts as "normal" time at each zone? Above what does it become an exception?
- Decide the alerting tier. Each zone: who gets notified, on what channel (email, Slack, SMS, customer webhook), and at what threshold?
- Build escalation paths. What happens at the 30-min, 90-min, 4-hour mark?
- Set false-positive guardrails. Valet mode for consumer, gate-queue debouncing for ports, GPS-jitter buffers everywhere.
- Audit weekly for the first month. Which alerts produced action? Which got ignored? Tune the rules; turn off the ones that don't earn their keep.
What to ask a vendor about geofencing
Most vendor demos show you a pretty map and a circle. Ask the questions that reveal whether layer 3 actually exists:
- Can a single geofence have different alert rules per vehicle group (e.g., this customer's containers vs that customer's)?
- Can I combine signals — entered zone and dwelling beyond X minutes and after business hours?
- Can geofence events fire webhooks to my own systems (TMS, ERP, customer portal)?
- How does the platform handle valet/exception modes — can a driver mute a single trip without disabling the whole geofence?
- Do you have built-in detention reporting from geofence dwell times, or do I have to write my own queries?
- What's the typical alert volume from existing customers running this scale of fleet, and what's the actionable rate?
A vendor who answers those clearly is worth your time. A vendor who pivots back to "we have unlimited geofences" is selling you the boundary, not the logic.
We rebuilt geofencing inside ViaLoop Fleet last year specifically because most platforms — including some we admire — get layer 3 wrong. If you're evaluating geofencing for a fleet of any size and want a 30-minute walkthrough of how we handle dwell-time reporting, route-corridor compliance, customer milestone webhooks, and false-positive suppression, the demo link at the top of the page goes straight to our team.