A mock recall is a practice run of your real recall plan: you pick a product lot, then trace it backward to its ingredients and forward to every customer who received it, timing yourself against a clock. The goal is to prove you can account for 100% of a lot's disposition fast, usually within two to four hours, before a real contamination event forces the issue.
Nobody wants to run their first recall live. A mock recall is how you find the broken link in your trace chain on a calm Tuesday instead of at 2 a.m. during an outbreak investigation. This post walks the whole exercise end to end: how to set it up, the trace-back and trace-forward legs, how to score it, and what the common GFSI schemes actually expect.
What is a mock recall?
A mock recall simulates pulling a product from the market without actually recalling anything. You choose a finished-product lot code, reconstruct exactly what went into it and where every case went, and reconcile the quantities. Auditors treat it as evidence that your recall plan works in practice, not just on paper.
It is different from a traceability exercise, though the two overlap. A traceability exercise tests whether your records connect one step up and one step down. A mock recall adds the management wrapper: assembling the recall team, drafting mock notifications, calculating the percentage of product accounted for, and often practicing the communication tree. Many plants fold a traceability test into a mock recall so one exercise satisfies both.
What are the two legs of a trace: back and forward?
Every mock recall runs in two directions from the lot you pick. Trace-back answers "what went into this?" You start with the finished lot and walk backward through the batch record to the raw materials, their supplier lot codes, and receiving dates. Trace-forward answers "where did it go?" You start with the same finished lot and walk forward through packing, warehousing, and shipping records to every customer, distributor, and ship-to address that received a case.
The forward leg is what a real recall depends on, because it tells you who to call. The backward leg is what an outbreak investigation depends on, because it tells you whether a suspect ingredient touched other lots. A complete mock recall does both, and reconciles the mass balance: pounds in should equal pounds shipped plus pounds on hand plus documented waste.
How do you run a mock recall end to end?
Run it like the real thing, on the clock, with the people who would actually do the work. Here is the sequence.
- Pick the lot and note the start time. Choose a finished-product lot at random, or have your quality lead pick one you have not drilled before. Write down the exact minute you start. The clock is the point.
- Assemble the recall team. Notify whoever sits on your recall committee, quality, operations, shipping, and a decision-maker who can authorize a hold. In a drill, confirm they can be reached and know their role.
- Trace back to raw materials. Pull the batch or production record for the lot. List every ingredient, its supplier lot code, and receiving date. This is where broken links show up: a hand-written mix sheet that does not record which sugar lot went in.
- Trace forward to customers. Pull shipping and warehouse records. Identify every case produced under that lot and every customer, distributor, and ship-to that received one.
- Reconcile the mass balance. Add up what you made, what you shipped, what is still on hand, and what was scrapped. The numbers should close. A gap means product is unaccounted for, the single most common finding.
- Draft the mock notifications. Write the customer notification and, if your plan requires it, a mock FDA Reportable Food Registry entry. You are testing whether you have the contact list and the template ready.
- Stop the clock and record the time. Note the total elapsed time and the percentage of product accounted for.
- Score, document, and fix. Debrief the team, write up gaps, and open a corrective action for each broken link. File the whole package as audit evidence.
How fast do you have to be?
The number auditors and customers expect is hours, not days. Under the BRCGS Food Safety standard, traceability must be demonstrated within four hours (or 24 hours when the information has to come from an external party), and the standard requires a traceability test at least every six months plus a full recall test annually. SQF requires a traceability exercise at least annually but does not specify a timed limit in the standard itself.
The tighter benchmark you will hear, 100% trace within two hours, usually comes from customers, not the schemes. Large retailers such as Costco set a two-hour expectation in their supplier requirements. So the honest answer is: BRCGS says four hours, some customers say two, and a good target is to beat both.
By the numbers. FDA maintains a public list of food recalls, market withdrawals, and safety alerts and keeps each posting available for three years (FDA, Recalls of Foods & Dietary Supplements). Under the FSMA Food Traceability Rule, covered firms must provide required traceability records to FDA within 24 hours of a request (21 CFR Part 1, Subpart S). A mock recall is how you find out whether you can actually hit those clocks.
How do you score a mock recall?
Score it on three numbers and one judgment. The three numbers: elapsed time, percentage of product accounted for, and mass-balance closure (pounds in versus pounds out). A passing mock recall accounts for 100% of the lot and closes the mass balance within a small tolerance for documented waste. The judgment is qualitative: did the team know their roles, were the templates ready, did the records connect without a phone call to someone's memory?
Anything short of 100% accounted for is a fail worth a corrective action, even if you finished under time. If you can trace fast but only reach 94% of the cases, the missing 6% is exactly what would hurt you in a real event. Treat every gap as a traceability defect and fix the record that caused it.
Why does the trace usually break?
It breaks where a record depends on memory or handwriting. The classic failure points: a mix sheet that records the ingredient but not the supplier lot, a rework step where two lots merged without a new lot code, a warehouse pick that shipped mixed lots on one pallet, or a shipping log that lists the customer but not the ship-to. Every one of those is a place where paper let a link go unrecorded.
This is where capturing data at the station instead of on a clipboard changes the drill. When operators record the actual supplier lot as they scan it into the batch, and shipping records tie each case to a ship-to at pick time, the trace-forward and trace-back become a query instead of an archaeology project. Harmony's traceability approach and its connected data model exist to make that lot linkage automatic, so a two-hour trace stops depending on which supervisor is on shift. That same live data is what makes a mock recall a five-minute report rather than an afternoon.
How often should you run one?
At minimum, match your scheme: BRCGS wants a traceability test every six months and a full recall test annually; SQF wants at least an annual exercise. In practice, run a mock recall at least twice a year and vary the scenario, different products, an ingredient-driven trace-back one time, a customer-driven trace-forward the next. Also run one after any major change: a new line, a new warehouse management setup, or a new co-packer. Practicing the same easy lot every year proves nothing.
How does a mock recall fit the rest of your food-safety system?
It is the stress test for three programs at once. It validates your recall plan your traceability records, and the discipline behind your HACCP program. It proves your team's readiness. A clean mock recall is one of the strongest pieces of evidence you can put in front of a GFSI auditor, because it shows the paper system survives contact with a clock. Build it into your food-safety calendar the same way you schedule internal audits and management review.