Drone Stockpile Measurement: The Complete Buyer's Guide

Buying a stockpile survey looks simple from the outside: someone flies the yard, a number comes back, the number goes in the inventory report. The reason it goes wrong is that the number is not the deliverable. The defensibility of the number is.

Piles get reconciled against sales records. Volumes get reviewed by auditors, by a parent company, or by a buyer during a transaction. When a number is questioned, what you need is not a bigger number or a prettier map — it is a written account of how the volume was arrived at that stands up when someone reads it closely.

This guide covers what actually drives accuracy, what actually drives cost, how long it takes, which deliverables to insist on, and the questions that separate vendors. It is written for the person signing the purchase order, not for a survey crew.

What you are actually buying

A drone stockpile survey is four things stacked together, and vendors differ on all four while quoting them as one line item:

1. A flight that captures the site with enough overlap and resolution to reconstruct it.
2. Survey control that ties the reconstruction to real-world coordinates.
3. Processing into a measurable surface and a point cloud.
4. A report that states each pile's volume, the material, and the base surface used.

The cheapest quotes usually thin out items two and four. That is not automatically wrong — a fast progress check does not need what a year-end inventory needs — but it should be a decision you made, not one made for you.

What actually drives accuracy

Accuracy conversations usually start with the aircraft. That is the least important variable.

Survey control

Control is what ties the model to the ground. It comes either from ground control points surveyed on site, from RTK or PPK positioning of the aircraft, or — on work that has to be defensible — from RTK/PPK with independent checkpoints used to verify the result rather than to build it.

That last distinction is the one that matters for audit. A survey that uses every control point to construct the model has no independent evidence of its own accuracy. A survey that holds a few points back as checks can state what the error actually was. If your number will be reviewed, ask for the check.

Ground sample distance

GSD is how much ground each pixel covers, and it follows from flying altitude and sensor. Finer GSD resolves the toe of a pile and the crest more sharply, which is where volume error concentrates. Flying higher is faster and cheaper and gives you a coarser model — another trade-off that should be explicit in the quote.

The base surface

This is the big one, and it is not a technical detail — it is the definition of the number.

A pile volume is always measured relative to something: a surveyed pad, a modelled base plane interpolated under the pile, or a previous survey's surface. Two entirely honest surveys of the same pile can report different volumes because they used different bases. A volume with no stated base is a number without a definition.

If your yard has a hard pad that was surveyed once, that is the strongest base and it should be reused every survey so your numbers are comparable over time. If the material sits on ground that moves, the base has to be modelled, and the report should say so.

The material itself

Two things a drone cannot see. It cannot see through a pile, so a void, a buried structure or a pile built over old material is not detected. And it measures volume, not tonnes — the conversion to tonnage uses a density factor that comes from your operation, not from the survey. If a report gives you tonnes, ask which density was applied and where it came from.

What drives cost

Quotes vary more than buyers expect, and the spread is rarely flight time. The drivers, in rough order:

• Site size and pile count — more area is more flight and more processing, but pile count drives the reporting effort, which is often the larger share.
• Control requirements — establishing and surveying ground control is field time. Reusing existing site control is much cheaper than establishing it, which is why the first survey of a yard usually costs more than the ones after it.
• Access and coordination — live operations, orientation requirements, escort, and airspace coordination all consume the day.
• Reporting depth — a volume table is one thing; a reconciliation against your records with period-over-period movement is another.
• Turnaround — compressed delivery costs more, everywhere, always.
• Travel — distance and whether the site can be combined with others in the same trip.

The practical implication: get quotes on an identical written scope, or you are comparing different products. If one quote is dramatically lower, look at control and reporting before assuming you found a bargain. We publish indicative ranges on our pricing page so the first conversation can start from something real.

Turnaround, and when to fly

Field time on a typical yard is a fraction of a day. Processing and reporting are the longer part, and honest turnaround for a standard inventory report is measured in days rather than hours.

The bigger scheduling question is when to fly. For inventory that gets reconciled, fly at a consistent point in your cycle — the same week each month or quarter — because comparability across surveys is worth more than any single survey's precision. Fly when the yard is as static as it can reasonably be, and tell the operator which piles are actively being worked so the report can note it.

Alberta adds a seasonal factor. Snow cover changes what the survey sees, and a pile under snow is not the same measurement as a pile without it. That is worth planning around rather than discovering in January; our note on winter drone operations covers what changes.

The deliverables to insist on

A stockpile survey should hand you all of the following. If a quote does not name them, ask.

• A volume table — each pile identified by the same name your operation uses, with its volume.
• A stated base surface for every volume, and a note where the base was modelled rather than surveyed.
• A stated coordinate system and vertical datum. Deliverables in the wrong projection are not a quick fix.
• An accuracy statement — what control was used and, ideally, what the independent checkpoints showed.
• The orthomosaic, so you have a dated measurable image of the yard.
• The point cloud and surface model, so a later question can be answered without re-flying.
• Survey date and conditions, including anything that affected the capture.

The point cloud matters more than buyers realise. It is the record that lets someone re-derive the volume against a different base a year later, when the original question has changed. Our walkthrough of what each survey deliverable is goes through the formats in detail.

Choosing a vendor

Five questions, in the order they discriminate best:

1. "What base surface will you use for each pile, and how will you report it?" A vendor who answers this fluently has done audited work. A vendor who has not thought about it has not.
2. "How will you demonstrate the accuracy of the result?" Look for independent checkpoints, not a specification sheet quoting the aircraft's positioning accuracy. Those are different claims.
3. "Are you prequalified in our contractor management system today?" Avetta, ISNetworld or ComplyWorks. This is the item that most often derails a mobilisation, and it cannot be fixed on the day.
4. "What are the deliverables, in what formats, in what coordinate system?" Get it in writing before the flight, because the deliverable list changes how the site is flown.
5. "What can't you do on our site?" The most informative question in the set. A vendor who names a limitation — vegetation, active faces, a pile that cannot be flown safely — is giving you something you can plan around. One who answers yes to everything has told you nothing.

Our contractor vetting guide covers the certification, insurance and data-handling side of the same conversation.

The five ways it goes wrong

• No stated base surface. The number cannot be defended or compared to the next one.
• Bases that change between surveys. Period-over-period movement becomes meaningless.
• Pile names that do not match your records. Reconciliation turns into a manual mapping exercise every cycle.
• Tonnage reported without the density factor stated. Nobody downstream can check the conversion.
• No point cloud retained. The next question requires a new flight.

All five are avoidable in the scoping conversation, and none of them are visible in a quote that only lists a price and a date.

Is drone the right method at all?

Usually, for stockpiles — the geometry suits it, the safety case is strong, and the whole yard gets captured rather than a sample. But not always. Heavy vegetation, very small pile counts, and sites where a ground crew is already mobilised for other work all change the calculation, and there are situations where a hybrid approach is the right answer.

We set the two methods side by side in our drone vs ground stockpile survey comparison, including where ground survey still wins.

Where to start

If you are buying this for the first time, do three things before requesting quotes. Write down the pile names your operation actually uses. Find out whether your yard has existing surveyed control. Decide who downstream consumes the number — inventory, finance, or an auditor — because that determines how much of this guide applies to you.

Then send the same written scope to every vendor. Talk to us if you want help writing that scope, including the parts where we would tell you a different method fits your site better.