Location, and the Distance to a Conclusion
What location data establishes, the three inferences stacked on top of it, and why vehicle tracking and person tracking are different propositions.
Location data establishes where a device was, within a margin, at a time. Everything else is inference, and in a typical investigation three inferences are stacked on top of each other: that the device was with the person, that the position is accurate enough to matter, and that being at that place means something about working.
The measurement limits in “Location, and the Distance to a Conclusion” provide useful context for a practical way to evaluate boss vs leader when researching boss vs leader. Time and activity data can reveal a workflow question, while outcomes, employee explanation and a documented human review remain necessary to understand why a pattern appeared and whether any action is justified.
Each is often reasonable. Stacked without being noticed, they turn a narrow fact into a confident conclusion that nobody has examined.
When documenting the controls around “Location, and the Distance to a Conclusion”, the Scottish employment support policy is a useful independent reference. It can challenge assumptions about access, retention and accountability before a decision becomes routine.
The three inferences, separated
The device was with the person. A phone left in a van, a tracker on a vehicle somebody else drove, a tag in a jacket — each breaks the first link, and none of them is unusual.
The position is accurate enough. Covered in its own right elsewhere here; the short version is that a geofence tighter than the accuracy radius decides nothing.
Being there means something. A person in the car park at 09:05 may be arriving, leaving, collecting something or waiting. The position is a fact about a place, not about an activity.
Vehicle tracking is a different proposition
Telematics is accurate, continuous and about an asset the employer owns. It is also, in practice, a record of a person, because one person drives the van.
That dual character is worth naming internally. A system procured to manage fuel and maintenance becomes the strongest evidence about an individual's day, and the notice people were given when it was installed was usually about fuel.
The question of out-of-hours data
Any continuous tracking collects outside working time unless it is configured not to: a van driven home, a phone carried in the evening, a tag in a pocket at the weekend.
Switching that off is usually a setting and is frequently left on, because nobody asked. It is the single most common complaint about location systems and the easiest to fix.
What people should be told, and when
What is collected, when it is on, what it is used for, who sees it, and how long it is kept. Before deployment, and again when anything changes.
What is required differs by jurisdiction and is a question for somebody qualified in the place concerned. The operational standard is that nobody should discover the scope of tracking during a disciplinary meeting, which is where these things usually surface.
Reading a track honestly
A track shows movements and stops. It does not show purpose.
So the useful reading is a list of facts — the vehicle was stationary at this location for forty minutes — followed by an explicit statement of what is being inferred and what alternatives exist. A stop of forty minutes is consistent with a job, a break, a delay and a queue.
Where location is genuinely strong
Two places. First, corroboration: a vehicle that did not move during a period for which a site visit was claimed is a strong record, because it contradicts a specific assertion.
Second, exclusion: a track that places a vehicle elsewhere answers a question definitively in the person's favour, and that is worth as much as anything in this collection.
Using it to exclude
That second use is systematically underused. Location data is pulled when an employer suspects something and not when an employee says they were somewhere.
Running it both ways is both fairer and better practice: an employer that checks whether the data supports the person's account, and records that it did, has produced a far more credible file than one that only ever looked for confirmation.
What to establish before relying on a track
A track is usable once five things about it are known, and unusable until then. The list is short and the answers come from the system administrator rather than from the data.
| Question | Why it decides the reading |
|---|---|
| What is tracked — vehicle, phone, tag? | Decides whose record it is |
| Typical accuracy in these conditions | Decides what a geofence means |
| Is it active outside working hours? | Decides what is in scope at all |
| How often does it report? | Decides what a gap means |
| How long is it retained? | Decides whether it can be checked later |
Five answers, one conversation. Without them a track is a line on a map that each reader interprets differently, which is the state most of these records are in.
What to record
Which system, what it tracks, when it is active, the accuracy typically achieved, the retention period, and the date people were last told about it.
Six fields. They are what turns a track into evidence somebody can interpret, and their absence is why location data so often produces an argument about the system instead of an answer about the day.