Document control without the bureaucracy

Document control has a reputation for ceremony: forms, stamps, signatures, all in service of itself. Stripped of ceremony, it answers one question. When a procedure was followed last Tuesday, which version of it was followed, and how does anyone know? Everything else in the discipline — version numbers, effective dates, issuance and retirement — exists to make that one question answerable months later, by someone who was not there.

What control actually controls

The risk is not an unapproved document. It is two versions of the same procedure circulating at once. A person is trained on the new version while the bench still uses the old, or a printed copy saved in a notebook in a drawer outlives its own retirement. The work then cannot be traced to a method, because the method is uncertain. Control is the set of practices that keeps one version current and all others visibly retired — no more, and it does not need to be more.

Four fields that carry the weight

A controlled document needs an owner, a version, an effective date, and an approval. Each field answers one question: who maintains this, which revision this is, when it applies, and who accepted it. An effective date matters more than most people expect, because a version number alone cannot settle what was in force on a given day. Where the approver is a role rather than a person — the responsible scientist, the quality lead — the role should be named; a signature line for “management” answers less with every reorganization.

Change history belongs with the document. A short table — version, date, what changed, why — turns every future audit question about the procedure into a lookup. It also makes the current version credible: a document that has never changed in ten years is either admirably stable or quietly unreviewed, and the history distinguishes the two.

Issuance and retirement

Controlled copies belong at the point of use. Where the master is electronic, a single source with a visible version footer is enough; where paper copies circulate, each one should be marked as controlled, and uncontrolled printing should be stamped or dated as a courtesy to its future reader. When a document is superseded, the old version comes out of circulation — but into an archive, not a bin. The version that governed a batch is part of that batch’s record, and a reader reconstructing the work in two years needs to find it. Superseded is not the same as deleted.

Review, proportionately

Documents have review dates, and a review may honestly conclude that nothing needs changing. The record of that conclusion is the point. Changes of any size route through the same approval as the original — an edit by the author alone is a draft, however small the edit. Proportionality does the rest: a small laboratory may control five documents, a large one hundreds, and the count measures nothing. What matters is that the set is defined, each has an owner, and the version in use can be identified on request.

The audit question

An assessor picks up a procedure and asks for its history: who approved it, when it took effect, what changed since it was written. If those answers live in the document’s own footer and change record, the conversation is short, and the laboratory looks like the kind of place that knows which version it used. That is all document control has ever promised — and delivered properly, it is enough.

Proportionality deserves its own sentence, because the failure mode in practice is mimicry. A laboratory that copies a larger operation’s document system inherits its ceremony without its staff, and the copy usually collapses into unsigned free-for-alls or unusable binders. Five documents, honestly controlled, are worth more than five hundred controlled in name.