The walk is over and there is a list. This is the part of an inspection that most changes what happens next, and it is where the difference between a restaurant that improves and one that gets steadily worse reports shows up — not in what was found, but in what was done about it in the ten minutes after it was found.
⚖️ Not all findings weigh the same
Every system in the world splits findings roughly the same way, whatever it calls them.
- The ones that can make somebody ill now. Food at the wrong temperature, cross-contamination happening in front of them, no hot water at the hand sink, a sick person handling food, chemicals stored over food. These usually have to be corrected immediately, they weigh the most, and they are the ones that bring somebody back.
- The ones that make the first kind more likely over time. A damaged surface, a missing label, a cleaning schedule not signed for three weeks, a gap under a door. These usually come with a deadline rather than an immediate correction.
The useful consequence: when you get the list, sort it by that split before you do anything else. The instinct is to start with whatever is quickest, and the quickest items are almost always in the second group.
🔧 Correcting it while they are standing there
A great deal of what appears on a report can be fixed on the spot, and in most systems a finding corrected during the visit is recorded as corrected — which is a materially different thing from the same finding left open.
So when something is raised and it can be dealt with now, deal with it now: move the product to the right shelf, discard what has to go, get the soap refilled, put the thermometer in and take the reading. Say what you are doing while you do it.
Two things not to do, though, and they matter:
- Do not fix it silently and hope it goes unrecorded. It was still found, and appearing to conceal costs more than the finding.
- Do not promise a fix you have not thought through. "We'll get that replaced this week" said under pressure becomes a missed commitment on the re-inspection, and a missed commitment is much worse than an honest "that one needs a contractor, I will get a date".
And when something cannot be corrected on the spot — most building and equipment issues cannot — say so plainly and ask two questions: what the deadline is, and what evidence they will want that it was done. Asking is not weakness; it is the single most useful minute of the whole visit, because it converts a vague finding into a task with a date and an acceptance criterion.
📄 The report, and reading it properly
Before anybody leaves, go through the report with them and make sure you understand each item — not whether you agree, whether you understand. A finding you misread is a finding you will fix wrongly and be shown again.
Three things to establish before the door closes:
- What exactly was observed, in enough detail that you could point at it. "Improper storage" is not actionable; "raw chicken on the shelf above prepared salad in the walk-in" is.
- What the deadline is for each item, since they are usually not all the same.
- Whether there will be a re-inspection, and whether you need to notify them when something is done or wait for them to come.
Take your copy, and if a score or grade came out of it, understand how it was arrived at rather than just the number. In systems where the grade goes in the window, the arithmetic matters commercially as well as legally, and it is usually driven by a small number of heavily weighted items rather than by the length of the list.
🙅 If you think they are wrong
Sometimes an inspector is working from a misunderstanding — a process they have not seen before, a piece of equipment doing something not obvious, a product that is not what it appears to be.
The way to handle it is the same in almost every jurisdiction. Explain once, calmly, at the time — with evidence if you have it in reach. If it is not resolved, do not press: note it, take the report, and use the formal route afterwards. Every authority has one, it is usually written on the report itself, and it usually has a short window.
What makes that route work is evidence that already existed: the logs, the plan, the supplier specification, the self-inspection lists. What does not work is a recollection. This is the third time in this run that the same sentence has been true, and it is not a coincidence — an argument you can only make with your memory is an argument you have already lost.