Multi-site audit scheduling: planning sampled sites under IAF MD 1
How to plan multi-site audit scheduling under IAF MD 1: sampled sites, the central function, rotating samples across the cycle, and building routes and teams.
Key takeaways
- IAF MD 1 lets certification bodies sample sites rather than visit every one, when conditions are met.
- Sample sizes start from the square root of the number of sites and are adjusted for risk.
- The central function is audited every time; sampled sites should change across the cycle.
- The scheduling challenge is routes, teams and dates across many locations at once.
Multi-site audit scheduling means planning audits for organisations with many sites under a sampling approach set by IAF MD 1. The central function is audited at each audit, a sample of sites is visited, the sample changes across the cycle, and every sampled site needs a competent auditor on a date and route that works.
What makes multi-site audits different
A retailer with 80 stores, a food group with 12 factories or a service company with 30 branches can be certified under one certificate, with a central function managing the system. Visiting every site at every audit would be impractical, so IAF MD 1 sets out when and how a certification body may audit a sample instead.
For the planner, that turns one audit into a small project: which sites, in which order, with which auditors, on which days, and how to change the sample next time.
The part of a multi-site organisation that manages and controls the management system, and is audited at every audit under IAF MD 1.
Which organisations can be sampled
Not every multi-site organisation qualifies. IAF MD 1 expects a single management system under a central function that has the authority to require corrective action at every site, and sites carrying out substantially similar activities. Where sites do very different work, they may need to be grouped and sampled separately, or audited individually.
Getting eligibility right before planning saves rework: an audit programme built on sampling that doesn't apply has to be rebuilt.
How sampling shapes the plan
| Audit | Starting sample | Notes |
|---|---|---|
| Initial certification | Square root of the number of sites | Rounded up; increased for higher risk |
| Surveillance | Around 0.6 × the square root | Can be lower where the system is proven |
| Recertification | Around 0.8 × the square root | Covers sites not recently visited |
These are starting points. IAF MD 1 requires adjustments for risk, complexity and past performance, and part of each sample should be chosen at random. Always follow your own procedures and the current edition of the document.
The scheduling challenges
Choosing and rotating the sample
Sampled sites should vary across the cycle, so every site is seen over time.
Routes across many locations
Visiting several sites in one trip, in the right order, with the fewest kilometres.
Teams for different processes
Different sites may need different technical competence, such as a factory and a warehouse.
Keeping the central function first
The central function audit often needs to come first or last, depending on your method.
A worked example
- Month 1Sample chosenA client with 36 sites: a sample of 6 sites plus the central function, part chosen at random.
- Month 2Team formedA lead auditor for the central function and a second auditor with food-processing competence for two sites.
- Month 3Route builtSites grouped into two trips of three sites each, keeping travel short.
- Month 3Dates confirmedEach site confirms a date inside the audit window.
- Next yearSample rotatedSix different sites chosen, favouring sites not yet visited.
Each step involves rules that a spreadsheet can't enforce: rotation of the sample, competence per site, windows and travel. That's where audit scheduling software earns its keep.
Common mistakes
MythThe same convenient sites can be sampled every year.
RealitySamples should rotate and include a random element, so every site is seen over the cycle.
MythAny available auditor can visit any site.
RealitySites may run different processes needing different technical competence.
MythSites can be booked one by one as dates come in.
RealityBooking them as a programme lets you cluster trips and keep the central function in the right place.
Planning tips
- ✓Keep a site register with each site's processes, location and last audit date.
- ✓Record which sites were sampled at each audit so the next sample can rotate.
- ✓Plan the central function audit and sampled sites as one programme, not separate bookings.
- ✓Cluster sampled sites into trips before assigning auditors.
- ✓Check that each auditor is competent for the processes at the sites they visit.
See how ScheduleAI's audit scheduling software applies these rules across a whole programme in minutes.
Can remote audits reduce travel for multi-site clients?
IAF MD 4 allows information and communication technology to be used where risks are assessed and controlled. Some site activities can be audited remotely, which changes the routing problem.
How do integrated audits interact with sampling?
Integrated management systems, covered by IAF MD 11, may need teams that cover several standards at sampled sites. See audit team selection.
ScheduleAI plans multi-site programmes as one job: it rotates samples across the cycle, forms teams for each site's processes, clusters sites into trips and keeps every visit inside its window.
Book a demo Estimate your savingsQuestions
What is multi-site audit scheduling?
Planning audits for organisations with many sites under a sampling approach, including the central function and a sample of sites.
What is IAF MD 1?
The IAF mandatory document that sets requirements for auditing and certifying multi-site organisations using sampling.
How many sites need to be audited?
Sample sizes start from the square root of the number of sites for initial certification, with reductions for surveillance and adjustments for risk.
Does the sample change each year?
It should. Sampled sites are rotated across the cycle so all sites are covered over time.