What is geotagging: geotagging explained, and what the coordinates actually claim
Geotagging is the practice of attaching geographic coordinates to a file such as a photograph, video or social post, normally recorded automatically by a smartphone at the moment of capture and stored in the file's location metadata fields.
For anyone assessing a claim or an inspection, one sentence governs everything that follows. A geotag records where a device reported being, which is not the same statement as where the subject of the photograph is.
How does geotagging work?
The phone establishes a position, then writes it into the file. The position comes from a mix of sources, not from satellites alone:
- Global navigation satellite systems, GPS along with GLONASS, Galileo and BeiDou, accurate to a few metres outdoors with a clear sky.
- Wi-Fi positioning, matching nearby network identifiers against a database of known locations. This is what usually works indoors.
- Cell tower triangulation, coarse and sometimes wrong by hundreds of metres.
- A cached last-known position, used when nothing better is available, which can be minutes or kilometres out of date.
The coordinates end up in the GPS fields defined by the Exif specification, maintained by the Camera and Imaging Products Association as CIPA DC-008 and developed with JEITA. Nothing in the file distinguishes a satellite fix from a cached guess.
What geotagging does not tell you
Four gaps matter in evidence work, and each has caught out a claims team somewhere:
- Device position, not subject position. A photograph of a vehicle taken from across a car park is tagged where the photographer stood.
- Capture position, not incident position. A car damaged in Rotterdam and photographed a week later on a driveway in Utrecht carries Utrecht coordinates, correctly.
- No accuracy statement. The fields hold a coordinate, not a confidence radius. A value good to three metres and one good to eight hundred look identical.
- Nothing about who held the phone. Location is not identity.
How geotags are spoofed
Location data is among the easiest evidence to falsify, and the methods need no technical skill. Coordinates in a file can be rewritten directly with free metadata editors. Mobile operating systems support mock location providers intended for developers, and consumer apps built on them will report any chosen position to the camera. Some tools generate an image with fabricated coordinates from the outset.
Detection is possible in places. Real-time capture inside a controlled flow, jailbreak and mock-location detection, and cross-checks against network-derived position all narrow the gap. None of them closes it. Anyone claiming location can be verified beyond doubt is selling something.
Geotagging explained: a practical example
A property claim covers storm damage to a roof. The submitted photographs carry coordinates matching the insured address, which reads as confirmation that the right building was photographed.
It is weaker than it looks. The coordinates confirm that a device reported being at that address when the file was written, and both halves of that sentence can be manufactured. What genuinely strengthens the position is triangulation across independent sources: coordinates recorded by the receiving system rather than read from the file, a server-side receipt time, network-derived location, and visible landmarks in the images consistent with the address. Agreement across sources the sender does not control is the useful signal.
When is geotagging worth requesting?
It earns its place as one input among several rather than as a control on its own. It is genuinely useful for confirming that a field engineer attended the right site, for grouping submissions from a large loss event, for spotting a batch of claims whose photographs all originate from one improbable location, and for flagging a mismatch worth a phone call.
It is not suited to being the deciding factor in a payment decision, nor to automatic rejection. A mismatch is a reason to ask a question, and there are innocent explanations for most of them. Fraud detection sits with people who can weigh context, as the wider argument in AI-generated insurance fraud and the layered approach described under evidence integrity both set out. Location alongside a digital fingerprint and a documented provenance record is worth far more than location alone.
Venta Capture, a product of VentaVid, records GPS location verification to within roughly 11 metres as part of its control points, captures in real time rather than accepting gallery uploads, and includes jailbreak detection. These are graded signals for a human reviewer to weigh, never an automatic verdict on a claim.