What an auditor is actually asking for

The Nine-Month Test

Jake Cooke-Wilkinson, Contracts Manager

Auditors do not read pest records forwards.

They tend to open the most recent inspection, then work backwards through the year, following findings, recommendations and recurring activity as they go. It is an efficient way to review a file, and it is the reason weak records come apart in a particular order. The recent reports are almost always tidy. The damage sits eight or nine months back, where a finding was raised and then quietly stopped being mentioned.

Attendance is the easiest thing in the file to prove. There will be a date, a signature and a report confirming somebody came. It is also the least interesting part of the document, because it answers a question nobody is asking.

What is actually being tested is whether the site can demonstrate control: that activity was identified, understood, acted upon and brought to a conclusion, and that the process held together across the whole review period rather than in the last quarter.

A finding has to survive the gap between visits

Take a real sequence. A technician records damage beneath a loading bay door in March and notes it as a possible rodent entry point.

By the time an auditor reaches the April report, they are looking for evidence that the finding stayed alive. Whether the repair was completed. Whether it was assigned to a named person. Whether an interim control went in while the repair was pending. Whether the technician returned to that area and said so.

A finding that disappears from the record without explanation is not read as resolved. It is read as lost. That distinction is worth sitting with, because in a great many cases the work was actually done. Somebody welded a plate over the gap in April and nobody wrote it down. The defect is gone from the building and still open in the file, and only one of those is auditable.

Isolated activity is often insignificant. The same observation appearing across several visits, without escalation and without resolution, is evidence that the programme is not reaching the underlying cause. What connects the two readings is continuity: whether the record preserves the thread from one inspection to the next, or restarts every time somebody attends.

Beginnings are well documented. Endings are not.

Most pest records are competent at the opening. Observations, photographs, recommendations, technician comments. The weakness appears afterwards, and it is almost always the same set of absences: no named owner, no target date, no status, no evidence that the work was completed.

A recommendation without an owner is a suggestion. A recommendation without closure can sit in a file for a year while everyone involved assumes somebody else has it.

The most persuasive document a site can produce is rarely the longest service report. It is a corrective action register that has been kept properly. Each entry should carry:

• what was identified
• the priority assigned to it
• the person responsible
• the action required
• the current status
• the completion date
• the evidence required to verify closure

Kept honestly, that register gives the programme a memory, and it stops the same recommendation being raised four times without anything moving. Where an action cannot be completed quickly, the entry should still record what interim measure is in place and when the permanent correction is expected. Nobody expects a structural repair to happen overnight. What is hard to defend is a material risk left open because responsibility for it was never formally taken.

The documents have to agree with one another

Pest records are not assessed as a stack of separate reports. They are assessed against each other, and the relationships between them are where inconsistency shows.

Trend data showing repeated mouse activity beside a corrective action register with almost nothing in it does not describe a controlled site. It describes two documents that have not been introduced. A site plan that has drifted out of step with the device register raises the same doubt, as does a risk assessment describing an operation the site no longer runs.

The same logic applies to proportionality. A low-risk office may hold a short file and run an entirely suitable programme. A food production site, care facility or complex distribution operation will require considerably more, because the consequences of failure are larger. Volume is not the measure. Fit is.

That is why quantity of equipment can work against a site. Thirty monitoring points that nobody can account for are less convincing than twelve placed for reasons somebody can articulate. Frequent attendance alongside structural defects that have been open for months reads as activity rather than control.

A well-organised folder can still expose a poorly connected programme. A plain one can be persuasive when the information is consistent, current and traceable. The objective is not to overwhelm a reviewer. It is to leave them without a question.

The test itself

Before any external audit, run this.

Choose one meaningful finding from roughly nine months ago. Not a recent one, and not a trivial one. Start at the original service report, locate the corresponding corrective action, identify who owned it, establish what happened at the next inspection, find the evidence of completion, and confirm the technician verified the correction and that the issue did not reappear unresolved later in the year.

That should take a few minutes and follow an obvious path.

When it does, the file demonstrates continuity, ownership and closure, which is the whole of what is being asked. When it is difficult, the information usually exists somewhere. The programme has simply failed to connect it, and an auditor can only assess what a site can demonstrate. The repair may have happened. The activity may have stopped. If the record cannot show it, the evidence is incomplete, and incomplete evidence is what a finding is written against.

Nine months is deliberate. It sits far enough back that nobody has rehearsed it, and inside the period an auditor will actually examine.

Stratton builds its reporting around that test under SEPOS, the Stratton Environmental Professional Operating Standard, a proprietary internal standard governing how pest risk is assessed, evidenced, corrected, reported and delivered. A finding is not closed because somebody attended again. It is closed because the record shows what changed.

Run the nine-month test on your own file before somebody else does. The result is more useful when there is still time to act on it.