How to Complete a Mock Food Recall in NZ and Account for Every Batch

A mock food recall gives a New Zealand food business a controlled way to check whether its records, people and procedures work together. The difficult part is often demonstrating where every affected batch went and explaining the quantity that remains unaccounted for.

This practical guide explains how to plan the exercise, follow batch records and complete a transparent stock reconciliation. It includes a worked example and a report checklist for manufacturers, importers and other businesses with relevant recall responsibilities.

What should a mock food recall test?

A mock food recall is a rehearsal of the decisions, records and communications needed when food may need to be removed from the supply chain. A traceability search is one part of it. Finding an invoice alone does not show that the team can assess a problem, identify all affected batches, prepare clear instructions and check whether the response worked.

Start with MPI’s simulated food recall guidance. Build a scenario relevant to your products and distribution, then test the response from the first warning through to a review of the exercise. A consumer-level scenario includes how you would communicate with consumers; a trade-level scenario tests removal from the food supply chain.

This article concentrates on the exercise and its evidence. For the wider context, see our guide to food product recalls in New Zealand.

Who needs an exercise, and how often?

MPI’s guidance describes simulated-recall requirements for businesses with plans or programmes under the Food Act, Wine Act or Animal Products Act, and for food importers and exporters. Run a simulated recall at least once every 12 months, measured from the previous simulated recall or a genuine recall that demonstrated effective traceability and recall procedures. An ineffective real recall does not qualify simply because it happened.

MPI makes a specific distinction for food service businesses selling only food for immediate consumption, such as a café serving meals. It does not expect them to have procedures for recalling that food from consumers or to run simulated recalls. They still need to respond to supplier recalls of ingredients and deal with food made from those ingredients. Businesses that also wholesale or sell packaged food need recall procedures and simulated recalls. Check the full activities and applicable plan rather than deciding from the business name.

For importers, MPI’s food importer recall requirements expressly cover regular simulated recalls and keeping recall records. If you are unsure which requirements apply, discuss your actual products and sales channels with your verifier or MPI. Record the basis for your decision.

Set up a realistic, controlled scenario

Choose a product that has a useful trail of real records. For example, an exercise could begin with a hypothetical supplier warning about an ingredient lot used in several finished batches. Alternatively, a simulated label error could affect a particular packing run. Label the invented trigger as exercise information; keep the actual production and distribution records separate.

• Define the boundary: name the product, pack size, lot or batch reference, relevant dates and the information initially available. Do not assume only one batch is affected before checking the records.
• Choose participants: include the people who would make decisions, retrieve records, control stock and prepare communications. In a small business, one person may hold several roles.
• Assign an exercise lead: agree who provides scenario updates, records decisions and prevents simulated instructions being mistaken for real ones.
• Set a cut-off: record the date and time used for the stock position so counts and transactions refer to the same snapshot.
• Define what you will assess: affected-batch identification, location and quantities, record retrieval, decision-making, draft communications and follow-up actions. Record elapsed times against your applicable procedure or customer requirements.

Trace backwards, forwards and across linked batches

Use the records to establish the affected scope before calculating percentages. A perfectly balanced total for the wrong batch is not a useful result. Show how each relevant ingredient or finished-product reference connects to a supplier, production run, storage location and destination.

• Backwards: connect the selected finished batch to ingredient lots, supplier deliveries and relevant packaging or labels.
• Forwards: identify every finished batch made with the affected input, then trace those batches to dispatches, stock locations and customers.
• Across the process: review split lots, repacking, relabelling, mixed pallets, work in progress and rework where these exist. Explain each transfer between identifiers.
• Beyond normal invoices: check internet orders, promotional supplies, staff sales, samples, exports, off-site storage and third-party logistics where relevant.

Keep a batch map or register with the record reference supporting every link. Distinguish confirmed information from an assumption or unanswered question. If a record is missing, describe the gap and its effect on the exercise; do not manufacture a historic entry.

Test how staff would identify and separate affected stock, prevent further use or dispatch, and recognise the correct hold status. MPI’s simulation guidance does not require a real hold on every simulated affected item. Record what was actually tested and what was discussed only.

Account for quantities without counting stock twice

Reconciliation compares the quantity in scope with the quantity supported by records. Pick one unit, such as jars or kilograms, and state the starting boundary. For a finished-pack exercise, use the number of finished packs produced or received. Do not subtract an earlier processing loss again if it was already excluded from that starting total.

At the agreed cut-off, each unit belongs in one mutually exclusive category. For example, count stock still held by your business separately from stock dispatched outside it. When investigating a customer’s portion, split that portion into the customer’s stock, onward sales and other supported outcomes; do not add those figures to the original dispatch total again.

On a small screen, swipe sideways to view all columns.

For bulk ingredients or rework, a count of retail packs may be insufficient. Use an appropriate mass basis, document conversions and follow the material into the receiving batches. Explain expected processing changes separately. Never describe an unexplained difference as “normal loss” just to close the exercise.

Worked example: 1,200 jars and a 30-jar gap

At the chosen cut-off, the team assembles the following non-overlapping evidence:

On a small screen, swipe sideways to view all columns.

Accounted for: 300 + 600 + 240 + 12 + 18 = 1,170 jars.
Unexplained difference: 1,200 − 1,170 = 30 jars.
Record-supported proportion: 1,170 ÷ 1,200 × 100 = 97.5%.

The team cannot report that all product has been accounted for. It records the 30-jar gap and checks transfers, dispatches and other supplies. In this example, it finds an existing dispatch record for 30 promotional jars supplied to customer C that was omitted from the initial search. Adding that distinct supply resolves the arithmetic difference: 1,200 jars are now accounted for.

That result establishes destinations and other recorded outcomes at this stage. It does not prove that every jar could be recovered, that no consumer received it, or that recall communications would work. Customers A, B and C still belong in the exercise’s communication and distribution assessment. Record why the promotional supply was missed and how the routine system will be improved.

If no reliable record explains the difference, leave it unresolved in the report and assign follow-up. Do not invent a sale, a sample entry or a disposal record. An honest gap is more useful than a false 100% result.

Rehearse decisions and prepare draft communications

Work through the scenario’s risk assessment and proposed recall response with the people responsible for those decisions. Record what information they used, what remained uncertain and why the simulated response was appropriate. The scenario needs to exercise a simulated trade-level or consumer-level recall. If the risk assessment instead concludes that no further action is needed, MPI says to undertake another simulated recall. Use the current forms and examples on MPI’s food recall documents page where helpful.

Prepare the draft information that would be needed for your verifier or New Zealand Food Safety, affected businesses and, for a consumer-level scenario, consumers. Check product names, pack sizes, batch or date identifiers, the described problem, proposed actions and contact details for consistency. Have someone compare the draft message with the batch map rather than checking the wording alone.

During the simulation, retain these as drafts. Do not send exercise notices to customers, publish recall advertisements or submit a hypothetical recall as a real event. MPI provides specific adaptations for simulated recalls, including draft regulator communications. Review whether the contact list and approval route would work without creating an actual incident notification.

Keep an exercise report that another person can follow

The report should let a reviewer reconstruct the exercise without relying on the memory of its organiser. Separate the scenario facts, real records checked, simulated actions and genuine gaps. MPI’s documents page includes scenario, checklist, affected-product, distribution and reconciliation resources that can help organise this evidence.

• Scenario, scope, start time, cut-off time and participants with their roles.
• Affected products and batches, and the explanation linking them to the hypothetical problem.
• Source records and a distribution list, with locations, quantities and relevant recipient details.
• Reconciliation method, units, calculations, assumptions and outstanding differences.
• Simulated risk assessment, decisions, hold-process test and draft communications.
• Timeline of key actions, including delays and unavailable information.
• Findings, corrective actions, responsible people, due dates and follow-up evidence.

Keep the records available for verification or other applicable review. Do not label a discussion as a physical stock check or a draft message as a sent notice. Protect commercially sensitive customer information when sharing the exercise file.

Turn findings into improvements and retest weak points

A useful exercise often identifies a weakness. The value comes from addressing its cause and checking that the correction works. A new form alone does not demonstrate that staff can complete it or retrieve it during an incident.

• Missing promotional supplies: include these in the normal dispatch controls, then trace a later promotional supply through the revised process.
• Mixed units: define the pack-to-carton conversion and check it against actual pack configuration.
• Unclear batch links: improve the production or repacking record and test the connection on another affected-batch scenario.
• Unavailable decision-maker: clarify deputies and decision responsibilities, then rehearse the handover.
• Outdated contacts: correct the contact register and define how it will be maintained.

Record what worked as well as what failed. Agree a completion date for each action and a proportionate follow-up check. Do not turn a percentage or an internally chosen time target into a universal MPI pass mark. The exercise needs to demonstrate effective procedures across its relevant scope.

How a food consultant can help you prepare

Access Food Verification & Auditing Limited can review traceability and recall procedures, examine records for gaps and help plan practical improvements. A clearly scoped discussion with a food consultant can help identify what to test and how to organise the resulting evidence.

Our food safety audit preparation support may help with reviewing recall-test records and follow-up actions. If a genuine product problem needs investigation, our product issue investigation support is a separate service to discuss. Consulting does not guarantee a verification or audit outcome; formal services remain subject to scope, recognition and independence requirements.

Frequently asked questions

Start with a scenario and a clear evidence trail

Select a realistic scenario, identify the affected scope and follow the records to each destination. Reconcile quantities honestly, rehearse the decisions and communications, then correct the gaps the exercise exposes.

General information only: Requirements depend on the complete food activities, products, processes, locations and sales model. Use current official guidance and confirm the requirements applying to the specific business before relying on a classification, timeframe or record list.