Nobody plans to do pointless work. We do it anyway.

I woke up this morning with a full day of things to do. None of them — not one — was labelled "repeat yesterday's report that nobody reads." And yet, somewhere between nine and ten, there I was.

That is the question worth sitting with: who in the history of the working world ever opened a calendar and wrote "10:00 — do the thing we've always done, even though we stopped knowing why"? Nobody. And yet.

**Why it happens — the honest version**

The simplest reason is habit. A process was designed at some point by someone with a real problem to solve. Then the problem changed, or the person left, or the tool that made the process necessary was retired — and the process stayed. It has momentum. Stopping it requires a decision, and decisions require energy. Continuing requires nothing at all.

The second reason is that busywork *feels* like work. A full inbox processed, a status report sent, a meeting attended — these feel productive. They are measurable in a way that the hard work is not. The hard work is the decision with no clear right answer, the conversation nobody wants to have, the problem without a ticket number. The repetitive task is safer. It is completable. You can draw a line under it.

Then there are the metrics that reward activity over outcome. Hours logged. Reports filed. Meetings held. These are proxies — they were invented because the real thing was hard to measure. But proxies become targets, and once a proxy is the target, you will get the proxy whether or not you get the thing it was meant to track. The report nobody reads still gets filed because there is a field in the system for it. The status meeting still runs because it is on the calendar.

And there is the comfort of the familiar task. I know how to do this. I have done it before. I can do it right now, without thinking too hard, and have something to show for it by noon. That is not laziness — that is the brain conserving energy by reaching for the known. It is only a problem when the known thing has stopped being necessary.

Copy-paste-rename. The spreadsheet that became three spreadsheets. The weekly summary that restates the dashboard. The approval chain built for a risk that no longer exists. Recognise any of these? Most people do.

**The turn**

Here is the thing about all of it: the part that is senseless is almost always the part that *repeats*.

The decision nobody wants to make does not repeat in quite the same way. The conversation that is genuinely hard takes a different shape each time. The problem without a ticket number is always slightly different from the one before it. These do not go in a pipeline. They belong to people.

But the report that is identical every Monday? The data transformation that runs the same way every time it is triggered? The status update assembled from three sources by hand, copied into a template, and emailed to the same list? That belongs in a pipeline.

I build the systems that remove repetitive work. Not because the people doing it are not capable — they usually are, and they usually know it is senseless, and they do it anyway because the alternative is a decision. The pipeline removes the need for that decision. The report runs itself. The transformation happens automatically. The status is assembled without a human in the loop.

What that gives back is not "productivity" in the abstract. It is hours. Specifically, the hours that were being spent on the thing that did not need a human. Those hours are now available for the thing that does.

**What automation is not**

It is not a replacement. The report that runs itself does not replace the judgement about what the report should contain — that still takes a person. The data transformation that runs automatically does not replace the decision about what to do with the result — that still takes a person. Automation removes the repetition. It does not remove the thinking.

And not everything repeats. Genuine judgement calls do not run on a schedule. Novel problems do not have a training set. The edge cases are the interesting cases, and they are the ones that most need a human.

The honest limit: I can only automate what I can describe. If the process is "whatever feels right today," a pipeline cannot help you. But if the process is "the same thing, reliably, every time this happens" — that is exactly what should not be done by hand.

**The plain version**

Nobody plans to do pointless work. But nobody questions it either, because questioning it takes more energy than doing it, and the calendar is already full.

The automation is not the point. The point is stopping the pointless thing, so there is room for the thing that is actually worth doing.

That is the whole argument. I did not need a report for it.