Your Process Documents Are a Record of Intentions

Outdated process documentation is not the result of neglect. It is the normal end state of every document that describes work, because a document records a decision made at a point in time and the work carries on changing after the file is saved. The question worth asking about your process folder is not whether it is current. It is which parts of it were ever true, and which stopped being true without anyone noticing.

Those are different failures and they need different responses.

Three ways a document stops describing reality

The first is drift. The process was accurate when written and has changed a little every few months since. A tool got swapped. A step moved. An approval threshold went up. No single change was worth a documentation update, and after two years the accumulated difference is large. Nobody can point to when it broke because it never broke, it eroded.

The second is that it was aspirational from the start. Somebody wrote how the process ought to run, either for a certification, a tender, or a new-hire pack, and the business never actually adopted it. This document has never been true. It is a proposal wearing the costume of a record.

The third is local divergence. The document is still accurate for one team and not for another. Two offices, two client types, or simply two people who were trained differently. Everyone reading the document assumes it is universal, and it describes a minority case.

Drift you can correct by asking. Aspiration you have to delete, which is politically harder. Divergence you have to split into two documents, which nobody wants to do.

The test that takes ten minutes

Pick any process document. Find the person who does the work. Read them the steps, one at a time, and ask a single question after each: is that what happens?

You do not need a workshop for this. You need ten minutes and a willingness to hear no.

The answers sort themselves into four piles quickly. Still true. Never was true. Was true, is not now. And a fourth pile that surprises people the first time they run this, which is steps the person does that are not in the document at all. That last pile is usually the largest.

Why the missing steps outnumber the wrong ones

Documents are written by people describing work, and description is lossy in a consistent direction.

Waiting does not get written down, because it does not feel like a step. Checking does not get written down, because it feels like part of doing rather than a thing in itself. Chasing does not get written down, because the person doing it does not count it as work. And workarounds do not get written down for the obvious reason that the document describes the route the workaround exists to avoid.

So a procedure that lists six steps might correspond to nineteen things the person actually does, and thirteen of them were never candidates for inclusion. The document is not wrong so much as it is a summary that has been mistaken for a specification.

What to do with a folder you no longer trust

Do not start by rewriting. Start by sorting, because most of the folder does not deserve the effort.

Rank the documents by how often the process runs and how many people it passes between. A weekly process crossing three teams is worth getting right. A quarterly one that a single person owns is worth a note, not a document.

Then, for the top few, run the ten minute test and rewrite only what fails. Rewriting from scratch is slower and produces a new document that will begin rotting on the same schedule.

If you would rather not maintain a folder at all, abi. Clone builds a free working model of how your business actually runs, with processes drawn against the real roles and systems they pass through, so a change to the business is visible as a change to the picture rather than as a file somebody forgot to open. Free on any account, no card, and a first pass takes minutes.

The uncomfortable defence of leaving it alone

Here is the argument against everything above, and it is a good one.

Documentation has a maintenance cost, and that cost is continuous while the benefit is occasional. If nobody has opened a document in a year, the honest response is not to update it, it is to accept that this process is carried in people's heads and to ask whether that is actually a problem. For a lot of small businesses it genuinely is not. The knowledge is distributed across enough people that no single departure is fatal, and the hours you would spend maintaining a folder are worth more spent elsewhere.

The exception is concentration. Where one person is the only route to a process, the absence of a document is a real risk with a real price, and that is the case worth spending on. Not the whole folder. The three processes with a single point of failure attached.

A stale document is worse than none

The reason to care at all is that people believe documents.

A new hire handed a two-year-old procedure will follow it, hit the parts that no longer work, and quietly conclude that they have misunderstood something. They will not conclude that the document is wrong, because documents carry authority. That costs them a fortnight and costs you their early confidence.

An empty folder produces a different behaviour. It sends them to ask a colleague, which is slower on paper and considerably faster in practice, and which has the side effect of surfacing the real process to somebody who has not yet stopped noticing it.

So if you are going to leave a document unmaintained, put a date on the top of it and let people judge it for themselves. That single line does more for accuracy than most rewrites.