What it means
A fleet dashboard shows no location for a van during a delivery shift. Fleet telematics data completeness asks what share of expected vehicle-time records were usable for the declared purpose.
It does not tell you whether every missing point was a device failure or whether drivers performed well, and Geotab's descriptions of device connection status and fleet data quality are examples of data pipelines, not evidence that a particular vehicle was online. Define completeness for the actual purpose first, because GPS journey coverage, maintenance odometer readings and diagnostic events require different fields and frequencies, then select expected minutes, trips or messages as the unit.
Set active time, since parked, decommissioned or off-shift vehicles may not need the same reporting cadence, and count expected records fairly: if sample rates adapt when parked, do not assume a constant ping every second. Review software updates too, because a firmware rollout can change the sampling interval and therefore the denominator, and check trip segmentation, since a fragmented trip may create several short records rather than one complete journey.
Require usable values, because a timestamp with impossible coordinates should not count just because it is nonblank, a feed attributed to the wrong VIN can be worse than a missing feed, and device clock drift or a wrong time zone can misorder events. Detect duplicates, since replayed messages inflate apparent completeness, and do not count idle pings as trip data if the use case needs actual journey coverage.
Check odometer and diagnostics separately, because a frozen reading can appear complete while failing maintenance scheduling needs, and error codes may be absent on older vehicles even when GPS is healthy. Distinguish delayed data from missing data, because a buffered trip can arrive later, so measure real-time and eventually complete views separately, and keep the original data so later backfill does not erase a real-time availability incident.
Cellular coverage, device power, battery state and installation can interrupt uploads, tunnels, remote sites and indoor garages produce expected connectivity losses, and a device can store data locally while the analytics service remains blank, so log source and receipt times. Define quality thresholds, such as accepted GPS accuracy and maximum permitted age for the task.
Keep reason codes: no device, network gap, ingestion error, invalid data and planned exclusion should remain separate, and a user intentionally disconnecting a device under an approved policy has a different root cause from damage. Record service outages, because a central ingestion failure affects many vehicles at once and should not be mistaken for driver behaviour, and monitor emerging gaps, since a rising lag between event and ingestion can warn of an outage before complete silence.
Show missing intervals, because a single percentage can hide a complete blackout during a critical handoff, and track each repair with a work order and an owner rather than just an exception label. Segment by hardware, meaning vehicle model, firmware and installation cohort, and compare rural and urban coverage appropriately before ranking drivers or depots.
Audit the source by comparing raw device logs, ingestion timestamps and the dashboard for a sample gap, and keep a safety fallback, because critical operations need an alternate communication path when data is unavailable. Preserve privacy by applying lawful purpose, access and retention controls, and do not overclaim precision, since completeness and accuracy are separate; use the metric for reliable operations and maintenance decisions, not as standalone evidence about employee conduct.
In practice
Real-world examples.
Example
A van has valid GPS points through a scheduled shift and meets the defined reporting interval. The dispatch map shows a continuous trace. Every expected slot counts as complete.
Example
A cellular gap backfills later; it counts incomplete in real time but complete in the final view. The report shows both figures, so the live outage is not hidden by the later backfill.
Example
A feed reports the wrong vehicle ID and fails identity validation despite having coordinates. The records are excluded from the numerator and the device is flagged for repair, with an owner and a due date.
Formula
Calculation
Completeness = expected active vehicle-time slots with timely usable records / all expected active slots x 100. Report maximum gap and backfill separately.
Worked example. An invented fleet has 20 vehicles, each active for 10 hours on a given day and expected to report every 5 minutes, which is 12 slots an hour.
- Expected slots = 20 x 10 x 12 = 2,400.
- Slots with timely usable records = 2,160, so real-time completeness = 2,160 / 2,400 x 100 = 90%.
- Later backfill recovers 144 slots, so eventual completeness = (2,160 + 144) / 2,400 x 100 = 96%.
The longest single gap, 95 minutes on one van, is reported separately, because a 96% eventual figure would hide it.Case study
Seen in the real world.
This entirely fictional case follows Alder Couriers. A device update reduced upload frequency on several vans, leaving the dispatch map intermittent. The team compared event and receipt timestamps, repaired the configuration and marked the live dashboard gaps separately from data that arrived later.
Alder also noticed that the affected vans shared one firmware version, which it had not segmented in earlier reports. It added firmware as a reporting split and checked the sampling interval after each rollout. No real employee behaviour is inferred from these gaps.
Watch out
Common mistakes.
- Calling nonblank but invalid coordinates complete data.
- Using eventual backfill to hide a real-time outage.
- Comparing drivers without accounting for coverage and device differences.
Questions
People also ask.
Does a complete feed mean accurate location?
No. Accuracy and timeliness need separate checks.
Should parked vehicles be included?
Use the declared active-time and sampling rules.
What if data arrives tomorrow?
Report live availability and eventual completeness separately.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related