Ohno's categories are a checklist for looking at a process and noticing what is consuming effort without producing anything anyone would pay for. They were written for a factory and they transfer to office work almost unchanged, which is a slightly damning thing to discover.

The test throughout is the same: would the customer pay for this step if they could see it itemized?

The seven

Transport. Moving things between places or people. A document passed through four reviewers has been transported four times. Each handover is a chance to wait, to lose context, and to need a conversation that would have been unnecessary.

Inventory. Work sitting between steps. Half-finished projects, unreviewed drafts, an unworked pipeline. It consumes attention while producing nothing, and it sets wait time for everything behind it — see Little's law.

Motion. Movement within a step that does not advance it. Looking for a file, switching between five tools, re-reading a thread to recover where things stood.

Waiting. Work stalled because something else has not happened. The most common and the most invisible, because the person waiting is usually doing something else and so nobody records the delay.

Overproduction. Making more than is needed, or earlier than it is needed. A weekly report nobody reads. A feature nobody asked for. Ohno considered this the worst of the seven, because overproduction causes the others: it creates inventory, which requires transport, which produces waiting.

Overprocessing. More effort than the outcome requires. Three revisions where one would have been accepted. A document formatted for a presentation that will be read on a phone.

Defects. Work that has to be redone. Expensive not only for the rework but for the disruption: a defect found late pulls people off whatever they had moved on to.

An eighth is usually added now — unused talent, people prevented from contributing what they could by how the work is arranged. Not Ohno's, and a real category.

Why it works better than looking for inefficiency

Because "be more efficient" is not actionable and the seven are. Each one tells you what to look at and what would count as finding something.

It also cuts against the strongest instinct in the room, which is that a busy process is a working one. Every item on the list is somebody being busy. The person transporting a document is working. The person producing an unread report is working. Nothing on this list looks like idleness, which is exactly why it survives review after review.

The two ways it goes wrong

Using it on people. The categories describe activities, and activities are produced by how the work is arranged. A person doing three handovers is not the waste; the arrangement requiring three handovers is. A waste exercise that concludes with a headcount reduction has usually found symptoms and removed the capacity that would have fixed the cause.

Eliminating the wrong waiting. Some waiting is a buffer doing a job, and removing it without fixing what it was absorbing produces the failure it was protecting against. This is the same distinction that governs just-in-time, and it is why waste reduction and margin of safety are not in conflict — one removes activity that produces nothing, the other holds capacity against uncertainty.

Starting

Take one process that people complain about and write down every step, including every wait and every handover. Mark each against the seven. The list is usually longer than expected and the largest item is usually waiting, which nobody had been counting because nobody was idle during it.

Then look at what is producing the waiting rather than at the waiting itself. It is almost always batch size or a handover that exists for a reason nobody remembers.