Frozen Food Traceability: A Mock Recall and Mass Balance Checklist

A frozen food mock recall should test whether the responsible people can identify the selected goods, follow their records backward and forward, reconcile the quantities and confirm the actions needed to control affected stock. For each location, establish whether the goods have been held, recovered or otherwise dealt with.

Set the exercise scope and stock cutoff first. Use the actual product, pack and lot references, retain the records retrieved during the test and show every unresolved link or quantity difference. Keep the calculation separate from the control result: “accounted for” describes an explained quantity, while “held” or “recovered” needs evidence of the corresponding action.

We start traceability discussions with the identifiers the buyer can see on receipt and the records needed to investigate that order. The facility, cold-store operator, logistics parties and buyer may hold different parts of the chain. Test the handoffs between those parties as well as the records inside one company.

Define the exercise before starting the clock

Choose a current product and a specific scenario. The starting point might be a finished carton code, an incoming ingredient reference or an identified packing-material lot. State what the participants are expected to find from that starting point. A backward trace from a finished pack and a forward trace from an input challenge different parts of the record system.

Identify the exercise controller and the people responsible for product information, warehouse records, customer distribution and the final assessment. Agree which external partners will participate and how exercise communications will be clearly marked. A simulated notification should not be mistaken for an actual recall instruction by a customer or an operational team.

Set timing expectations from the requirements that apply to the business, customer and exercise. Record the actual elapsed time for meaningful milestones, such as locating the first batch record, completing the distribution list and obtaining the required warehouse response. A single completion time can hide a long delay in one essential handoff.

The FSANZ Food Industry Recall Protocol, which addresses recalls in Australia, recommends testing the recall plan through simulations based on current products. Set the frequency and scope appropriate to your applicable obligations and programme. There is no universal response-time or completion-percentage threshold supplied by this article.

If the exercise uncovers a real food safety concern, escalate it through the actual incident procedure immediately. Keep real incident decisions separate from simulated results. The responsible team should assess the concern and follow the notification and control requirements that apply to the goods and market.

Trace the finished lot back to its inputs

Begin with the physical identifier and check that the record system interprets it consistently. Capture the product name, pack format, lot or packing reference and relevant facility context. If a date appears on the carton, establish what the date means. A packing date, production date and date mark are not automatically interchangeable identifiers.

The GS1 Global Traceability Standard distinguishes product-level identification from batch or lot identification and describes recording links between process inputs and outputs. It also addresses links between products and larger handling units such as pallets. These distinctions explain why a product code alone cannot tell you which particular production lot was involved.

Follow the finished pack into its packing record, then into the relevant production and input records. Compare identifiers, quantities, dates and the parties that created them. Retain the connecting reference at each step. For each document, identify the reference that connects it to the selected carton or its inputs.

Where a process uses several input lots, identify all the lots included in the selected output. Where one input lot feeds several outputs, follow those outputs as well. Include relevant intermediate goods or rework in the review. The CFIA’s traceability inspection guidance, Element 7.1, specifically examines identification through processing, including intermediate and reworked material where applicable. Its regulatory context is Canadian; the practical linkage question also belongs in the agreed exercise scope.

Check packaging links when the scenario concerns a label or material issue. Identify the packaging reference used for the finished lot and the records that connect the material to its supplier. If a label version changed during the relevant packing period, ask which finished goods received each version. Do not narrow the affected scope to one carton code simply because that was the first code supplied.

For every missing connection, state what is known and what remains uncertain. Have the assessor decide whether to widen the scope or request another record. Keep the original uncertainty visible even if a later response resolves it; it is evidence of how the system performed during the exercise.

Follow every outward movement

Build a lot-specific movement list from the relevant dispatch and storage records. Include the quantity, date, destination location and receiving business for each movement. Distinguish a transfer to another cold store from delivery to a customer. Commercial ownership and physical custody may sit with different parties, so identify who can locate or control the goods at each stage.

Two cartons of the same fruit can look alike while carrying different lot references. Use the recorded identifiers to select the shipment lines, then compare carton marks or label evidence where needed. Do not identify affected goods from a product photograph alone or assume that every item under the same product code belongs to the same lot.

Check split shipments, partial pallets and transfers between locations. If a pallet was broken down or rebuilt, find the record that connects its contents before and after that operation. Where several lots share one container, preserve the lot-to-quantity relationship inside the load. The container number identifies a shipment unit; it does not replace the identity of its contents.

Compare the distribution list with the underlying transactions and available receiving records. A commercial invoice may provide useful product and quantity information while leaving a lot relationship to another record. Identify that record explicitly. If the dispatch and receipt use different units or product references, document the conversion or mapping used to compare them.

Where a distributor has supplied further customers, define how the next stage of the trace will be demonstrated. Record whether the exercise includes a live partner response, a controlled simulation or only a review of the partner’s documented process. State exactly which response or record was tested.

The Australian recall protocol and supporting guidance describe the need to account for distribution and to consider onward distribution through intermediaries. For your exercise, show which receiving locations have been identified, which partners have responded and which downstream records are still outstanding.

Reconcile quantities without counting them twice

Choose the quantity basis before adding the records together. Define the lot or input being reconciled, the start and cutoff points, the unit and any conversion between packs, cases and net weight. Use mutually exclusive categories for the final position. Keep the transaction history beside the final stock position so each movement can be checked.

Consider a hypothetical finished lot, L1, containing 10,000 kg. At one agreed cutoff, the records show 3,000 kg in warehouse stock, 5,800 kg still with customers, 200 kg returned and segregated, and 900 kg with documented destruction. These categories account for 9,900 kg. The remaining 100 kg is unreconciled and needs investigation.

In this example, gross shipments to customers were 6,000 kg before the 200 kg return. Adding 6,000 kg shipped and 200 kg returned as separate final positions would count the returned goods twice. Use 5,800 kg with customers plus 200 kg returned for the cutoff position, while keeping the original dispatch and return transactions in the movement history.

The accounted proportion is 9,900 ÷ 10,000 × 100 = 99%. That is not a 99% recall rate, a pass threshold or confirmation that the customer stock is under control. Investigate the missing 100 kg and assess the control status separately. The figures are chosen to illustrate the calculation, not expected loss levels or an actual XMG exercise.

A production mass balance has a different boundary from this finished-lot example. It must connect the recorded inputs to outputs, remaining material and documented dispositions for the chosen process. Include applicable rework and any verified process adjustments, such as additions or removals, on a consistent basis. Do not force raw-material weight to equal packed-product weight by inserting an unexplained “normal loss” entry.

When a difference appears, check unit conversions, transaction timing, transfers, samples, returns and duplicated records before changing the total. Preserve the original calculation and show the evidence supporting each correction. A revised spreadsheet that balances only because someone changed the starting quantity does not resolve the underlying traceability question.

Test communication and stock controls

Separate the ability to send a message from confirmation that the right person received it and took the required action. For participating locations, identify the contact, an alternative contact and the person who can stop the affected goods from being dispatched or used. Test the handoff agreed for the exercise and retain the response.

The CFIA recall-procedure guide distinguishes notifying clients from verifying their actions and recording the result. It also describes identifying and segregating affected goods, reconciling quantities and monitoring returns. In a mock exercise, record which of those actions was actually demonstrated and which was only discussed or simulated.

A warehouse response should identify the lot, location, quantity and control status. If the exercise tests a dispatch block, obtain the relevant acknowledgement or system evidence within its agreed scope. A person saying that the stock is “in the freezer” confirms neither the selected lot nor the restriction on its movement.

For goods already with a customer, retain the customer’s quantity basis and the time of its response. Distinguish stock still available, goods further distributed and any returns or other dispositions reported. Where the exercise does not include that customer, mark the downstream control result as untested. Do not convert an internal distribution list into an assumed customer acknowledgement.

Include returns in the control plan. Identify who would receive the goods, preserve their lot identity, separate them from available stock and record the authorised disposition. Define how a return would update the customer and warehouse quantities without creating a second copy of the same goods in the reconciliation.

Keep exercise messages and screenshots clearly identified in the report. Record any operational restrictions agreed for the test, including whether a system block was demonstrated in a test environment or on selected live stock. State whether each control was applied, demonstrated in a test system or described without a demonstration.

Write an exercise report that exposes the gaps

Build the report from the records and responses produced during the exercise. Include the initial scenario, scope, participants, start and cutoff times, retrieved records, quantity calculation and control results. Show the points at which information arrived so a later reader can distinguish the first response from a corrected or completed response.

Use separate findings for record linkage, quantity reconciliation and control confirmation. Keep a missing customer list or an untested dispatch restriction open even when the backward trace and quantity calculation are complete. Give each finding a reference to the record, observation or response that supports it.

A concise result summary can report:

• Which input, finished-lot, storage and customer links were demonstrated.
• The quantity basis, amount accounted for and unexplained difference.
• Which participating locations confirmed the required controls.
• The actual time taken to reach the defined milestones.
• Missing records, inconsistent references and delayed responses.
• The corrective action, responsible person and planned verification for each gap.

Retain the source files used in the calculation with their relevant versions or retrieval times. If an inventory report changes as operations continue, the saved cutoff record explains the original result. Avoid relying on a live dashboard that will no longer show the same stock position when the exercise is reviewed later.

For information that cannot be shared broadly, agree a controlled review arrangement. Preserve the references needed to assess the trace while protecting unrelated supplier, customer or commercial details. An authorised reviewer may need access to supporting records even when the public or general supplier-qualification copy is redacted.

Correct the gaps and test the changed process

Turn each finding into a specific correction. If lot references disappear during a cold-store transfer, identify where the mapping should be created and who checks it. If a customer contact is outdated, update the contact arrangement and test the replacement. If returns are counted twice, correct the calculation method and verify it with another set of transactions.

Distinguish repairing the selected record from correcting the process that produced the gap. Reconstructing one missing link can finish an investigation while leaving the same weakness in tomorrow’s order. Ask what will make the required information available at the right stage without relying on one person’s memory.

The FSA’s current incident guidance for England, Northern Ireland and Wales describes using root cause analysis after a food safety incident to identify preventive actions. For exercise findings, establish why the gap occurred, assign the change and check whether it works. Follow the actual incident procedure if a real safety issue is involved.

Retest the changed part of the system with a suitable product, lot or movement scenario. Record the new result and link it to the original finding. Keep the first exercise outcome intact; a successful retest shows improvement, while the original report shows why the change was needed.

Review the plan when the operating route changes. A new facility, pack configuration, storage partner, software interface or customer distribution model can alter the identifiers and handoffs involved. Confirm which parts of the previous exercise still represent the current route and where a fresh check is needed.

Use the findings when agreeing the next order’s record requirements. Agree the identifiers, quantity basis and record owners needed to answer questions after receipt. Our traceability record framework connects those requirements with the proposed product, facility and shipment references.

References

The linked CFIA material has a Canadian regulatory context; the FSANZ protocol concerns Australia, and the linked FSA guidance states its UK jurisdictions. GS1 provides the cited identification and record-linkage framework. The L1 quantities and diagrams are hypothetical. They do not describe an actual recall, prescribe a universal acceptance threshold or establish that any particular shipment is safe.