The MapTrack Fleet Census, 2026 edition
The equipment fleets in this study do not specialise. Grouping every asset they track into eight classes, the median fleet spans five of them and only 10% confine themselves to one or two. Trade tools sit alongside heavy plant, vehicles alongside height-access gear and electrical test equipment. The second pattern is what decides which of those get tracked first: items that expire or must be inspected, not simply the ones worth the most.
This is a first-party study, and the distinction that matters most is how it was taken: the product findings are measured, not surveyed. Nobody was asked to recall their own practice. Every one of those figures comes from what equipment fleets had actually configured and recorded in MapTrack on 18 August 2026, across a census of every fleet meeting the qualification test rather than a sample of them. Findings are reported as a share of fleets rather than as pooled volume, so no single large customer drives a headline. Two findings draw instead on recorded conversations with equipment managers over the preceding 12 months, and say so where they appear. The population test, that second source and the limits are all stated in full at the end.
GM of Operations
A fleet that tracks excavators usually also tracks the drills, the ladders, the ute they travel in and the laptops in the office. Only 10% of fleets confine themselves to one or two classes; 60% span five or more. Any tool built for a single category is being asked to cover ground it was not designed for.
Source: MapTrack product data, August 2026. Shares rounded to the nearest 5 points.
Within that breadth, what gets tracked first is what carries a date. The categories appearing in the most fleets are dominated by items under an inspection or expiry regime: height-access gear in 60% of fleets, safety equipment in 50% and electrical test equipment in 50%, against heavy plant in 65% and vehicles in 60%. The ratio is what makes the point rather than the ranking. A harness costs a fraction of what an excavator costs, and a fleet is almost as likely to be tracking it. Nobody starts a tracking programme for height-access gear on replacement cost alone. What these categories have in common is that somebody has to be able to prove, on the day it is asked, that the item was checked and is still in date.
The categories above are one way to read intent. What fleets add to the asset record themselves is a second, independent one, and it points the same way.
Source: MapTrack product data, August 2026. Shares rounded to the nearest 5 points.
Date fields appear in 95% of fleets and attachment fields in 90%. Money fields appear in 60%. When an operator decides the standard asset record is not enough, the thing they most often add is a date to watch and a document to attach against it. That is a compliance register wearing the clothes of an asset register.
Full access to every feature. No credit card required. Per-asset pricing so you scale as your fleet grows.
Equipment tracking is usually sold against loss. Asked what prompted them to look for a system, the operators in 344 recorded conversations named something else first.
Source: MapTrack recorded conversations, August 2026
A failed or upcoming audit is named in 47% of those conversations, outgrowing a spreadsheet in 41%, and lost or stolen equipment in 22%. Read together rather than as a ranking, the picture is coherent: a deadline is what creates the urgency, and the reason the deadline is frightening is that nobody can quickly prove where the gear is or that it was checked. Loss and compliance are not competing motives. They are the same problem surfacing at the moment somebody has to produce evidence.
This matters because it is a third independent line of evidence for the same conclusion. What fleets configure, what they track and now what made them start looking all point at compliance rather than cost.
Fleets that run inspections are more than three times as likely to also raise work orders: 75% of them do, against 20% of fleets that do not inspect. The chain continues. Fleets raising work orders are twice as likely to run scheduled maintenance (70% against 35%), and those running schedules are more likely again to record meter readings and to track parts.
These are associations measured across fleets rather than a controlled comparison, so they describe which practices travel together rather than proving that one causes the next. What they describe is consistent and strong.
Read as a sequence rather than as a set of features, this is how a maintenance discipline actually starts. It does not begin with a maintenance system. It begins with somebody having to complete an inspection, finding a fault, and needing somewhere to put it. Every practice further down the chain follows from that one moment.
Across the measured population, 75% of assets in the median fleet have been checked out or transferred at least once. This is not a register sitting still. It is a working population of gear that changes hands, which is the difference between an asset list and an asset system.
How it changes hands is the more interesting half. Only 5% of movements pass directly from one person to another; the rest route through a location on the way. Custody is being handed back to a store, a site or a vehicle and then reissued, rather than quietly passed along a crew. That is the discipline an audit actually tests, and these fleets are keeping it at the moment it is easiest to skip.
Of every file attached across the measured fleets, 96% are photographs, and every fleet in the population attaches them. PDFs appear in 75% of fleets, documents in 20%, spreadsheets in 10% and video in 10%.
The more useful detail is where those files land. Photos attach to the asset record in 95% of fleets, but also to inspection forms in 60% and to check-outs in 70%. Evidence is being captured at the moment gear changes hands or gets inspected, not assembled afterwards into a folder. For anyone designing a field process, that is the behaviour to design around: the camera is the primary input device.
Among fleets running scheduled maintenance, roughly 60% of triggers are time-based and 40% meter-based. Of the time-based intervals, months dominate heavily, years are a distant second, and scheduling in days is rare.
Where maintenance is metered, engine hours and kilometres carry almost all of it, among the fleets that run metered maintenance. The tail is the interesting part: the measured units also include pressure, litres, cubic metres and temperature.
That is not a vehicle fleet with an odometer. It is mixed plant where a compressor, a pump and a truck each need a different basis for the same question. It is also the clearest practical consequence of the breadth finding above: a scheduling tool that only understands kilometres cannot serve a fleet like this, which is why those fleets end up back in a spreadsheet.
Sites are the dominant location type, present in 95% of fleets. The finding worth pausing on is what sits behind them: 40% of fleets model vehicles as locations, and by sheer number of location records vehicles rank second only to sites, ahead of dedicated storage.
A vehicle recorded as a place to hold gear, rather than as a vehicle on an asset list, means the fleet has accepted that its store room drives away every morning. Two in five of these operations track equipment on that basis, which is why the moment gear leaves a yard is the moment worth recording.
No operator was asked what they do. The product findings in this report are taken directly from what fleets had configured and recorded in the platform on the measurement date, which removes the two weaknesses that limit self-reported research: people misremembering their own practice, and only the most engaged organisations answering. What a fleet set up is a matter of record rather than recollection.
The trade-off runs the other way, and it is a real one. Measurement of this kind sees what was done but not why, and it sees only organisations already running a system. Both limits are set out below.
This study measures how equipment fleets currently work. It is not a description of what the MapTrack platform supports, and the two should not be read as the same thing. Operator practice moves at the speed of crews, procedures and audit cycles, so a practice that is uncommon here may be well established in the product and simply earlier in its adoption curve. Where a finding is scoped to the fleets doing a particular thing, that scoping describes those fleets, never the platform's capability.
This is a census rather than a sample. Every fleet meeting the qualification test is included, measured on 18 August 2026. The population is defined by that test rather than by a headline number. MapTrack does not publish absolute counts of its customer base, so no count of fleets and no band is given here or in the downloadable dataset. What travels with the data instead is the qualification test itself, which is the part that makes the measurement repeatable. The test is deliberately strict and behavioural rather than commercial: a fleet must have real users at their own email domain, at least 25 live assets and activity within the last 90 days. Trial, dormant, staff and demo accounts are excluded, which is what makes the figures describe working practice rather than sign-ups.
Classifying by how an account was created does not work on this data, and it is worth saying why. Accounts set up on a customer's behalf are owned by a MapTrack person, so a provenance test files real customers alongside internal ones. Behaviour separates them cleanly.
Two findings in this report do not come from product data. The trigger comparison and the compliance-versus-cost comparison are drawn from 344 recorded conversations with equipment managers over the 12 months to August 2026, analysed for topic presence rather than quoted.
The sample should be read for what it is: operators who were actively considering a change to how they manage equipment. It is offered as corroboration of findings measured independently in the configuration data, never as a standalone survey.
Pooled totals in a dataset like this are dominated by the largest few customers, which turns a finding about two organisations into something that looks like an industry statistic. Every headline here is therefore the proportion of fleets doing something, not the volume of it.
Shares of fleets are reported to the nearest 5 percentage points. A census of this size does not support a finer figure, and reporting one would imply a precision the measurement does not have. The two findings drawn on a large, already-published denominator, the 344 recorded conversations and the full population of attached files, are reported exactly. Every row in the downloadable dataset carries its own precision so the distinction travels with the data.
Fleets name their own asset categories, so the eight classes used here are a grouping applied afterwards, not a field anyone selected. Category names were collected across the population and restricted to names appearing in four or more independent fleets, which removes anything bespoke to a single organisation. Those names were then mapped to eight classes by meaning, and a fleet counts once in a class if any of its categories map to it, regardless of how many assets sit inside.
The mapping is a judgement, and it is the least mechanical step in the study. It is applied once and held fixed, so later editions group the same way and any movement is movement in the fleets.
Only observed events and existing configuration are counted. Two candidate findings were dropped because they rest on optional fields that are recorded too inconsistently to describe anything wider than the handful of fleets that maintain them.
This describes active operators, not the whole market. Every figure comes from organisations that have already committed to managing equipment systematically and are doing it daily. That makes it a useful picture of working practice, and not a survey of equipment operators generally.
There is no industry segmentation. Fleets are not grouped by sector in this study. Asset class prevalence is presented as what these fleets track, and should not be read as a market breakdown by industry.
It is a baseline. Configuration and cumulative activity as at 18 August 2026. Later editions can measure movement against it, but nothing here should be read as a trend yet.
Every figure is produced by a query held in version control against a written, fixed definition of the qualified population. That is what makes an annual series possible rather than a sequence of unrelated reports: the next edition re-runs the same measurement instead of approximating it, so a change between editions is a change in the fleets and not in the method. A figure whose measurement cannot be repeated exactly is not carried forward.
Version 2026.3, 18 August 2026. Shares of fleets are now reported to the nearest 5 percentage points, which is the precision a census of this size supports. Findings drawn on the 344 recorded conversations and on the full population of attached files are unchanged and remain exact. No figure was re-measured: this is a change in how precisely the figures are reported, not in what was measured. Two comparisons were restated so they stay true of the rounded values, and one comparison that had been wrong since first publication was corrected, since height-access gear is tracked by slightly fewer fleets than heavy plant rather than more.
Version 2026.2, 18 August 2026. Every figure was re-derived from committed queries and the report re-based to that measurement. Thirteen figures moved, most by a point or two. Four moved materially: the share of assets that have moved at least once, the share of movements passing directly between people, and the asset-class shares for vehicles and for heavy plant. The earlier values were produced by definitions that had not been written down, so they could not be reproduced exactly and were replaced rather than retained. Where a finding is worded differently as a result, it is because the measurement changed the finding, not the wording.
Version 2026.1, 17 August 2026. First publication.
All queries were aggregate-only and read-only: counts, distinct fleet counts and percentages. No customer is named or identifiable, no record contents were read, and asset category names are only reported where the same name appears across four or more independent fleets, so nothing bespoke to a single organisation can surface.
This report is free to quote, republish and build on under CC BY 4.0, including commercially, with attribution to MapTrack and a link to this page. No permission request needed.
The MapTrack Fleet Census 2026, MapTrack, https://www.maptrack.com/research/what-fleets-actually-track-2026
MapTrack. (2026). What Equipment Fleets Actually Track (2026). The MapTrack Fleet Census, 2026 edition [Data set]. https://www.maptrack.com/research/what-fleets-actually-track-2026
The underlying figures are available as CSV and JSON, each carrying the population definition, the measurement date and the attribution line. Every chart on this page has its own embed code. Questions about the method, or a correction, go to corrections@maptrack.com.
A wide mix, and that is the first finding. Grouping every tracked category into eight classes, the median fleet spans five of them: trade tools appear in 95% of fleets and fluid handling in 70%, followed by heavy plant at 65%, then IT equipment, height-access gear and vehicles level at 60% each, with electrical test equipment and safety gear at 50% each. Only 10% of fleets confine themselves to one or two kinds of thing. Within that breadth, the categories appearing in the most fleets are dominated by items under an inspection or expiry regime rather than simply the most valuable ones.
The configuration data points to compliance. When fleets extend the asset register with their own fields, date fields appear in 95% of them and attachment fields in 90%, while money fields appear in 60%. Combined with the categories that rank highest, which are dominated by inspection-regulated and expiry-dated items, the register is being used to answer "is this safe and current" more often than "what did this cost".
Roughly 60% of maintenance triggers are time-based and 40% are meter-based, among the fleets running scheduled maintenance. Of the time-based intervals, months are by far the most common. Where maintenance is metered, engine hours and kilometres carry almost all of it, though the measured units also include pressure, litres, cubic metres and temperature, which reflects genuinely mixed plant rather than a vehicle-only fleet.
Usually a deadline. Across 344 recorded conversations with equipment managers over the 12 months to August 2026, a failed or upcoming audit was named in 47%, outgrowing a spreadsheet in 41%, a safety incident in 23% and lost or stolen equipment in 22%. Read together these are not competing motives: a date on the calendar creates the urgency, and what makes that date difficult is being unable to quickly prove where equipment is or that it was checked. The sample is operators who were actively considering a change, so it describes what prompts a search rather than the prevalence of each problem.
With photographs, overwhelmingly. Of all files attached across the measured fleets, 96% are images, and every fleet in the population attaches them. PDFs appear in 75% of fleets, documents in 20% and spreadsheets in 10%. Photos are attached not only to the asset record but at the moment of handover and inspection, which suggests evidence is captured as work happens rather than filed afterwards.
Cite it as: The MapTrack Fleet Census 2026, MapTrack, https://www.maptrack.com/research/what-fleets-actually-track-2026. This is the 2026 edition, the first in an annual series. It is published under a Creative Commons CC BY 4.0 licence, so it is free to quote, republish and build on, including commercially, provided MapTrack is credited and the report is linked. No permission request is needed. The figures behind every chart are downloadable as CSV and JSON, each carrying the population definition, the measurement date and the attribution line, and every chart on the page has its own embed code.
No, and it should not be read that way. Every figure describes organisations that chose an asset tracking platform and are actively using it, so it is a study of a customer base rather than of the market. It also carries no industry segmentation, because no industry, sector or trade field exists in the underlying data. Asset category prevalence is offered as what these fleets track, not as a market breakdown.
Join construction, mining and field service teams across Australia.
4.9 on G2 ยท 5.0 on Capterra
