An inspection found a defect. What happens now?

Defects get found reasonably reliably. What is much less reliable is what happens in the hour afterwards, because that is where responsibility moves from the person who saw it to somebody who has to do something about it — and that handoff is usually informal.

The immediate decision comes first

Before anything is written down, one question has to be answered: can this structure stay in use as it is? That decision belongs to the inspector, it should be recorded explicitly, and it should have only three possible answers — available, restricted, or out of use. Leaving it implicit means the structure stays in use by default, which is the wrong default when nobody has actually decided.

Where a structure is restricted or taken out of use, the control has to be physical as well as documentary. A note in a report does not stop anyone climbing it; a removed or changed tag and a physical barrier do.

Then the work, with an owner and a date

A defect that has not become a job with a named owner and a date is still just an observation. The failure mode is familiar from every other trade: the report is emailed to someone who will raise the work when they get a moment, and a busy fortnight is enough for the defect to exist only inside a document.

Who else needs to know

A defect on a structure another trade is using is not only a scaffolding matter. The principal contractor, and often the specific trades working off that lift, need telling — and that notification should be recorded too, because "we told them" is a weak position compared with a record of who was told and when.

Common mistakes

• –Not recording an explicit decision on whether the structure may stay in use
• –Restricting a structure on paper without a physical control
• –Leaving defects inside a report with no job raised against them
• –Relying on one person to convert findings into instructed work
• –Not notifying the trades actually using the structure
• –Closing the finding when work is scheduled rather than when it is verified

In practice: the defect that stayed in an inbox

An inspector found a missing brace on a lift being used by a following trade, noted it, and emailed the report to the contracts manager at four on a Friday. The manager was on site the following week, the email sat unread, and the structure stayed in use for nine days with nobody chasing the work.

Everybody behaved reasonably. The inspector recorded it, the manager was busy, and no rule was broken. What was missing was any mechanism that made an open defect visible to somebody other than the person who happened to receive an email — which is a system failure rather than a personal one.

Repeat defects are telling you something else

A defect that recurs on the same structure is a different problem from a defect that appears once. A tie that needs attention at three consecutive inspections is not a maintenance item; it is a question about the substrate, the design assumption or the loading.

Recording defects against the structure rather than against the inspection is what makes that visible. Where each inspection is a separate document, three instances of the same defect look like three routine findings that were each dealt with properly.

Worth knowing: tell the client before they find out

A defect found and fixed is routine. A defect a client discovers before being told is a question about whether the contractor is on top of the job. Where a finding affects the client's programme or their other trades, saying so promptly costs nothing and buys a considerable amount of credibility.

• –Tell the client about anything affecting their programme or other trades
• –Record the notification, not just the defect
• –Follow up with confirmation of closure rather than waiting to be asked

Where this connects: repeats are a different problem

A defect that appears once is maintenance. The same defect on the same structure at three consecutive inspections is a question about the substrate, the loading or the design assumption — and it will not look like one if each inspection is a separate document, because three instances of routine maintenance is exactly what it resembles.

Recording defects against the structure rather than against the visit is what makes the pattern visible. It is the same argument as for ties and anchors, where a marginal result is information about the whole pattern rather than variation to absorb, and it needs the same thing: history held against the thing rather than the event.

• –Hold defects against the structure so repeats surface as patterns
• –Flag a third occurrence as a design question, not a maintenance one
• –Record the assessed cause, not only the symptom
• –Tell the trades using the structure, and record that you did
• –Make the restriction physical, not just documentary
• –Close on verification, with evidence attached