Asset loss: separating what walked from what was never where the record said

01 / FIELD NOTE

Keep the decision tied to the operating context.

An asset that cannot be found is usually described as lost, and the word collapses three different situations into one. The asset may have left the premises without authorisation, it may be somewhere on them in a place the record does not know about, or it may never have been where the record said in the first place. Each of those has a different cause and a different remedy, and a count that reports a single variance figure cannot tell them apart.

Misplacement is the largest of the three in most operations and the least discussed, because it is not a security problem and therefore attracts no attention. It is also indistinguishable from loss until someone looks: goods on the premises, in the wrong place, reported absent by a system that only knows quantities. The business buys replacements, which converts a location error into a purchasing cost, and then finds the originals, which creates a surplus nobody explains.

Record error is the second and is often the result of the first. A movement that was never posted, a receipt entered from paperwork that never arrived, a transfer created to clear an earlier discrepancy — each produces a record that is wrong without anything having been lost. This is the category that responds to process rather than to security, and it is where most of the improvement usually is.

Disappearance proper is what remains, and it is smaller than a variance figure suggests. Distinguishing it requires knowing which specific units are unaccounted for rather than how many, because a quantity tells you something is wrong and an identity tells you what and when. That is the difference between a stock adjustment and an investigation, and it is the practical reason unit-level identity earns its cost in an asset-tracking application.

The mechanism that makes this possible is the last recorded event. If the record holds the movement history of a specific asset — where it was seen, when, and in whose custody — then an asset that has stopped appearing can be placed at a point in time and a process. An asset that was last seen at a loading dock is a different question from one last seen in a service bay, and neither is answerable from a quantity.

Portals and gates matter for the portion that genuinely leaves, and their value is narrower than security marketing implies. A boundary read confirms that a tagged asset crossed a point at a time, which supports an investigation and does not prevent a removal. What it does well is close the question of whether an asset left the building, which is otherwise a matter of opinion, and that is worth having even though it stops nothing.

The timing of detection is the variable worth optimising, because it changes what can still be done. An asset missed at its expected return, noticed the same day, is a phone call. The same asset noticed at a quarterly count is a write-off. Frequent counting is what moves the detection earlier, and the reason to count frequently is not the accuracy of the count but the recoverability of what it finds.

Availability is the mirror requirement and is frequently conflated with loss. An asset that is present but reserved, awaiting inspection, or assigned to a job is not available, and treating availability as though it were presence produces over-purchasing that looks like a demand problem. The record needs to hold the state as well as the location, because the two answer different questions.

The physical realities decide what can be tracked at all. Metal equipment changes how a tag behaves and pushes toward tags designed for it; outdoor storage adds weather and impact; small accessories are the easiest to lose and the hardest to tag, which is why they are usually the first line on a shortage report. Those constraints should be identified during planning rather than discovered during an investigation.

Frequency has a cost and the cost is worth measuring rather than assuming. Counting everything often is expensive; counting the categories where loss is expensive, or where movement is frequent, or where an asset leaving would be consequential, concentrates the effort where it changes an outcome. A differentiated rhythm gives the same hours a better return than a uniform one.

The output that matters is not a variance figure but an exception queue with causes. Expected and not found, found but not expected, found in the wrong place — each of those is a different finding and each needs a different owner. A count whose discrepancies are all adjusted to a single account has converted a detection into a concealment, and the same discrepancy will recur at the next count.

The honest position is that none of this stops a determined removal. What it does is shrink the unexplained portion of a variance to the part that is genuinely unexplained, so that the effort spent on the problem goes where the problem is. That is a smaller claim than asset-loss prevention usually makes, and it is the one a record can actually support.