The scanner is a witness, not a judge

The system tells you what has been recorded. The manager has to understand what has happened.

In The system is faster than you I closed with a list of the essays to come, and one of them was about the scanner that is a witness and not a judge. This is that essay. About the device that, in last mile, usually has the last word, and about the question of what that word actually says.

What I have and what I do not

What I have: the work in last mile with loading problems, missing bags and mixed trolleys. The reports that also mentioned mismatches during scanning and loading. Incomplete and Damage as KPIs we followed every day. The rules of the operational instructions: when a bag was missing, the driver told the Team Leader and the trolley did not leave the designated zone, and chilled and frozen products were scanned inside the temperature-controlled zones. From the hotel, the difference between what the system records and the real state of a room, which I wrote about in the first essay.

What I do not have: a verified case in which the system showed a bag as scanned and I saw for myself that it was missing. The original document of the trolley rule, its date and its author. A case in which the scan history was used to blame a driver, or in which I used it myself to establish responsibility. A case in which the scanner was not working, and how the work went on. And from the hotel, no case in which a recorded status or transaction did not match reality.

So this essay is built on the problems and rules I had in front of me every day, and on what I would recommend today. There is no reconstructed incident in it.

What a scan proves

A scan is a short and precise sentence: this code was read, in this place, at this time. That is all.

It does not say that the bag is still there. It does not say that it was put on the right trolley. It does not say that the right trolley went to the right driver, nor that the driver took everything before leaving. All of these are steps that happen after the scan, in the physical world, and the scanner does not see them.

That does not make the scan less valuable. It makes clear what it is valuable for. A good witness says precisely what they saw. They do not say what happened when they were not there.

An accurate record is not always proof that the physical process ended correctly.

The trolley rule

In Where part of the order goes missing I wrote about the rule when a bag was missing: the driver told the Team Leader, and the trolley stayed inside the designated zone until the problem was solved.

I do not know who wrote that rule or when. But my reading of it is this: the rule does not ask anyone to look at the screen. It asks someone to stop and look at the trolley. At the moment when the record and reality may not match, the rule gives the word to the physical check, and keeps the trolley where that check can still be done.

That is my interpretation of the principle, not a quotation from a procedure. But it is the principle I hold to for this whole essay.

A witness, not a judge

When a bag is missing, the first question that comes up in most operations is not “where did it go missing”. It is “who did it”. And the scan history looks like the quickest answer: here is who scanned it last.

But a missing bag may have gone missing in the warehouse, during preparation, during loading, or on the road. The problems we had in last mile could involve any of these parts. The last scan shows who touched the bag last in the system. It does not show where the bag was separated from its place.

In KPIs do not improve in Excel I wrote that when you treat the number as the answer, you punish the driver for a bag that was never on their trolley. With the scan it is the same thing, one step closer.

That does not mean nobody is responsible. When it is proven that someone broke the standard, individual responsibility remains. But the proof comes from reconstructing the process, not from the name on the last line of the history.

The scanner is a witness to an action. Judging requires understanding the whole process.

When the scan is missing

There is also the other side: the scanner that does not work, the scan that is not made. I do not have a case to describe, and I do not know how it was officially regulated.

But one thing I would not suggest: that the work goes on without a scan when the procedure requires one. This applies especially to chilled and frozen products, where the scan inside the temperature zone is itself part of the rule. A missing witness is not replaced with an assumption. Every exception is handled according to the authorised rules, and written down.

Five steps for a mismatch

What I would do differently today is the way a mismatch between the system and reality is investigated. I have not applied it as a protocol in every case. But I would recommend it like this:

1. Check the record and its time. What exactly the system says, and when it said it.
2. Verify the physical state of the load. With your own eyes, not with another scan.
3. Find the last point where the information and reality matched. The mismatch was born after it, and that is where the search starts.
4. Separate the recording error from the preparation or loading error. They are two different problems, with two different solutions.
5. Document the confirmed cause and the measure that prevents it repeating. Confirmed, not assumed. And a measure, not just a name.

The third step is the one that is often skipped. Without it, the search starts at the last line of the history, and ends there.

At the front desk

In Front Office and Night Audit, the system records reservations, room statuses, check-ins, check-outs and other operational information. The complaints about water pressure and room cleanliness, which I have documented, required communication with the responsible departments, because the real state had to be verified where it was, not on the screen.

I do not have a case in which the system showed a room as in order and I found the opposite. I will not invent one to make the comparison prettier. The principle is the same as with the scanner, and I wrote it in the first essay: the status in the system and the full history of a problem are not the same thing.

A mismatch between the system and reality

The card below is for every case where the system says one thing and the trolley, the shelf or the room says another: what was recorded and when, what the physical state shows, where they last matched, whether it is a recording error or a loading error, and what the confirmed cause is, with its measure. Fill it in before you talk about blame. Incomplete Control keeps the cases side by side, so that a mismatch that repeats at the same point looks like a pattern and not like bad luck.

The system tells you what has been recorded. The manager has to understand what has happened. Between the two is the process, and that is where the work is done.