Allocating five audit days across six sites
You are planning a surveillance audit of a logistics company: head office and main hub (820 staff) plus five depots of 40–90 staff. You have two auditors and five audit days. Last year's programme audited all five depots, which consumed 60% of the time; four depots produced no nonconformities. Current data: on-time delivery 98.7%, up from 97.9%; damaged-goods complaints up 40% year on year; a new automated sorting system went live at the hub five months ago; night-shift turnover is 31%; two depots have opened since the last audit; one depot handles a single client's pharmaceutical distribution under a contract with tight temperature-control requirements.
1. Propose an allocation of the five days and justify it in risk terms.
Read the worked answer
Risk criteria first - change, complexity, performance signals, regulatory/contractual criticality, and audit history. Then allocate.
| Area | Time | Why |
|---|---|---|
| Head office + main hub (incl. sorting system) | ~2.5 days | Largest population, the new automation is a five-month-old process change, complaint trend sits here, most interfaces |
| Pharmaceutical depot | ~1 day | Regulatory and contractual criticality - the highest consequence of failure in the estate |
| Two new depots | ~1 day combined | No audit history; unknown control environment |
| Two established, clean depots | ~0.5 day, partly remote/documentary | Low change, good history - sample rather than visit fully |
Points to make explicit
- Equal allocation per site is a time-tabling habit, not a risk judgement. Last year 60% of effort bought almost no findings at four depots - that is the evidence for re-balancing.
- Newness is a risk factor even without performance data; a site with no history cannot be assumed low risk.
- The 40% complaint rise and 31% night-shift turnover both point at the hub and at shift handovers, which the plan should name as specific audit trails.
- Keep some float: if the sorting system turns out to be poorly controlled, the plan must be able to follow it rather than stick to the timetable.
2. On-time delivery improved while damage complaints rose 40%. How could both be true, and what would you audit?
Read the worked answer
Plausible relationships
- Metric definition: on-time measured at dispatch or depot arrival rather than at customer delivery. Damage discovered later never appears as lateness.
- Perverse incentive: pressure to hit the OTD target drives faster loading and less protective packing - the two indicators move in opposite directions because one is being optimised at the expense of the other.
- Process change: the new sorter may damage goods (mis-sorts, jams) while speeding throughput - consistent with a five-month-old system and a complaint rise building over the same period.
- Recording shift: complaints may simply be captured better now; the trend needs a data-quality check before it is believed.
What to audit
- The OTD definition, its data source and who can change it; whether management review questioned a rising OTD alongside rising damage.
- A consignment followed end to end: order, pick, sort, load, transport, delivery, complaint handling - the interfaces are where the two indicators collide.
- Sorter performance data: jam rates, damage at induction, maintenance and operator training since go-live.
- The complaints themselves: sample 10–15 and check whether corrective actions address handling, packing or the sorter.
3. The auditee's quality manager says: "Last year's auditor ran through the clauses in order and we were done in three days. Why are you following consignments?" Explain the difference and its value.
Read the worked answer
Clause-by-clause auditing walks the standard's table of contents. It is easy to plan and easy to defend, and it reliably produces findings that each clause has some artefact. It is poor at showing whether the system works, because the evidence is gathered out of sequence and interfaces fall between clauses.
Process-based auditing starts from the organisation's own processes - order-to-delivery, sortation, complaints, supplier management - and follows work through them. It tests inputs, outputs, sequence, interaction, resources, performance indicators and the risks the organisation has identified at the points where they actually occur.
Why it fits this client
- The interesting question (OTD up, damage up) sits across processes - sales commitments, hub operations, transport, complaints. No single clause owns it.
- The new sorter is a process change; only an end-to-end trail shows its knock-on effects.
- Following a consignment naturally surfaces handover points between shifts - relevant given 31% night-shift turnover.
Practical answer: the clause structure is still used as a coverage check at the end of planning and before closing, so nothing is missed. It is a map, not the route.