Your phone buzzes with a discount coupon the moment you walk past a coffee shop. A delivery app notifies the customer that the driver is two minutes away. A construction company gets an alert when a vehicle leaves the job site after hours. All of these are geofencing: virtual boundaries drawn on a map that trigger automatic actions when a device crosses them.

Geofencing sits at the intersection of GIS and everyday mobile life. If you work with maps, fleet data, or location-based services, it is one of the most practical geospatial concepts to understand - and one of the easiest to put to work.

A geofence is a virtual perimeter defined around a real-world geographic area. It can be a simple circle around a point - say, a 200-meter radius around a store - or a precise polygon tracing the exact footprint of a building, a campus, or a delivery zone. The boundary exists only as data, but software treats crossing it as a real event.

When a GPS-enabled device such as a phone, vehicle tracker, or wearable enters or exits the boundary, the system fires a pre-programmed action: a push notification, an automatic clock-in, an alert to a dispatcher, or a status update in a database. The fence itself does nothing; the intelligence lives in the rules attached to it.

Geofencing combines digital mapping with real-time positioning. A GIS specialist or developer draws the boundary using coordinates, then the system continuously compares device locations against it. Three positioning technologies do the heavy lifting, often together.

GPS provides the backbone outdoors, accurate to a few meters under open sky. Wi-Fi positioning takes over indoors, where satellite signals cannot reach, by triangulating against known access points in offices, malls, and hospitals. Cellular tower data fills the gaps with broader but less precise coverage, keeping tracking continuous when the other two drop out.

Not all geofences behave the same way. Active geofencing tracks location continuously in the background, giving high accuracy and instant triggers at the cost of battery life - the approach used by fleet management and workforce apps. Passive geofencing checks location only when another app requests it, which saves power but can miss quick boundary crossings.

Geofences also differ by trigger direction. Entry triggers fire when a device crosses into the zone, exit triggers when it leaves, and dwell triggers fire only after a device has stayed inside for a set time - useful for distinguishing someone walking past a store from someone actually shopping in it.

Fleet and asset tracking is the classic use case: dispatchers get automatic alerts when trucks enter depots, leave approved routes, or sit idle too long. Field service companies use geofences to clock technicians in automatically when they arrive at a job site, cutting timesheet fraud and paperwork.

Retail marketing made geofencing famous - the coupon-on-the-sidewalk trick - but the same pattern serves safety and compliance. Construction firms fence hazardous zones and get alerts if workers enter them. Drone operators use geofences to keep aircraft out of restricted airspace. Parents fence school routes and get notified when a child arrives safely.

These two terms get mixed up constantly. Geofencing triggers actions based on crossing a precise virtual boundary in real time. Geotargeting is broader and simpler: delivering content based on a general location such as a city, zip code, or neighborhood, with no boundary-crossing event involved.

A weather app showing your city's forecast is geotargeting. A parking app that starts your session the moment your car enters the garage is geofencing. The difference matters when you are choosing tools, because true geofencing needs continuous location monitoring while geotargeting usually needs just one lookup.

Geofencing only works ethically with user consent, and in many jurisdictions, legally too. Reputable implementations ask permission for location access, explain what the boundary is for, and aggregate or anonymize data rather than tracking identifiable individuals where possible. If you are building a geofencing feature, consent is not a checkbox to bury - it is the foundation the whole feature stands on.

For analysts working with geofence data, the same care applies on the back end. Store the minimum location history you need, set retention limits, and separate identity from movement data wherever the use case allows it.

The barrier to experimenting is low. Mobile operating systems ship native geofencing APIs that handle the battery-efficient monitoring for you, and spatial databases like PostGIS can evaluate boundary crossings with a single query. Define one small zone, attach one simple action - a log entry, not a customer notification - and watch it fire. From there, the path to production geofencing is mostly a matter of scaling up what already works.