A field supervisor pulls up the live map, sees an officer’s dot sitting on the wrong side of the building from where they’re supposed to be, and calls the post to find out what’s going on — only to hear the officer is exactly where they should be, standing at the front entrance. This happens often enough at every guard company running GPS tracking that it’s worth answering the question directly: how accurate is GPS tracking for security guards, actually, and why does the dot drift?
What a phone is actually doing to find itself
A phone doesn’t get a single, precise location fix the way people assume. It blends several sources — GPS and other satellite constellations, nearby Wi-Fi networks it can see, and cell tower triangulation — and the phone’s operating system decides which combination to trust based on signal strength and availability at that moment. Outdoors, in open sky, with a clean GPS signal, that blend is fairly tight. Indoors, near tall buildings, underground, or in dense urban cores where satellite signal bounces off structures before it reaches the phone, the blend leans harder on Wi-Fi and cell data, which are inherently coarser.
That’s the entire mechanism behind a dot that’s twenty or thirty feet off from where an officer is actually standing. It’s not a broken app, and it’s not the officer lying about their position. It’s the phone doing its best with degraded input and reporting a best estimate, not a lie.
Why “urban canyon” is the phrase to know
Tall buildings close together create what’s usually called an urban canyon — satellite signal reflects off glass and steel before it reaches the phone, and the device ends up averaging several bounced signals instead of one clean one. A downtown office tower post, a parking structure, or a site wedged between two high-rises will show more drift than a suburban strip mall with open sky overhead, and that difference has nothing to do with the officer or the software — it’s a property of the physical environment the post sits in.
Underground and structurally shielded locations are the more extreme version of the same problem. A loading dock enclosed on three sides, a basement mechanical room, an underground parking level — GPS satellite signal often can’t reach the phone at all in these spots, and the map will show the officer’s last known position from before they went in, or a rough estimate based on nearby Wi-Fi if any is available. That’s not a malfunction. It’s the physics of where radio signal can and can’t travel.
The phone’s own settings matter as much as the building
A meaningful share of the “the app is broken” complaints supervisors chase down turn out to be device settings, not app failures. A phone with battery-saver mode aggressively restricting background location access will report position less frequently and less precisely, because the operating system is deliberately trading accuracy for battery life. An officer who’s disabled location permissions for the app beyond “while using,” or who has Wi-Fi turned off entirely — removing one of the three inputs the phone would otherwise blend — will show worse drift than a phone with full permissions and Wi-Fi enabled, even standing in the exact same spot.
This is worth building into onboarding rather than discovering it one incident at a time: confirm location permissions are set to always-allow or the app’s required equivalent, confirm Wi-Fi is on even where the officer has no intention of connecting to a network (the phone can still use nearby networks it detects for positioning without joining them), and check battery-optimization settings aren’t quietly throttling the app in the background. None of that eliminates the physics-based drift near tall buildings or underground, but it removes the drift that’s actually just a misconfigured phone, which is a different problem with a much easier fix.
So, how accurate is GPS tracking for security guards in daily use
Don’t treat every drift as an incident. A dot that’s briefly off from the checkpoint location, on a site with tall structures or underground areas nearby, is expected behavior, not a red flag. Reacting to every instance trains supervisors to distrust the map generally, which is worse than the drift itself.
Do treat a sustained, unexplained gap as worth checking. The distinction that matters isn’t “is the dot slightly off” — it’s “has the phone stopped reporting a location update for an extended period with no environmental explanation.” That’s a different problem, and it’s the one worth a radio call.
Checkpoint scans matter more than the live dot for proof of presence. This is the core reason a guard tour system with physical checkpoints — NFC or QR — exists alongside GPS tracking rather than instead of it. A scan proves the officer’s phone was physically within range of a specific tag at a specific time. The live map dot is a general-position indicator with known limits, useful for real-time awareness, not the thing you’d put in front of a client as proof an officer was at an exact spot.
Set expectations with clients up front. A property manager who’s told in advance that GPS accuracy varies by building type won’t panic the first time they see a dot drift on the portal. A property manager who discovers this on their own, mid-contract, reads it as the software failing rather than as ordinary physics.
Where accuracy actually matters and where it doesn’t
For vehicle patrol across a large property or multiple sites, GPS accuracy in the tens of feet is more than sufficient — nobody needs sidewalk-level precision to confirm a patrol vehicle covered its route. For proving an officer stood at one specific door at one specific moment, GPS alone was never the right tool regardless of accuracy, which is exactly why checkpoint scanning exists as a separate, complementary layer rather than a redundant one.
The mistake worth avoiding in either direction is treating GPS as either useless because it drifts, or as precise enough to stand alone as proof of presence. It’s neither. It’s a general-position tool with known, physics-based limits, most useful for situational awareness and route confirmation, and it works best paired with the checkpoint-scan layer for anything that needs to hold up as documentation.
If your current setup has supervisors chasing every GPS drift as if it were a missed post, that’s a training gap on the tool, not a tool that needs replacing. Get in touch to see how live tracking and checkpoint verification work together without generating false alarms.