A recall is the worst possible time to discover your records have holes. A mock recall finds those holes on a quiet Tuesday instead: you pretend one lot is contaminated and time how long it takes to answer where it came from, where it went, and whether the numbers add up. Auditors expect it, certifiers require it, and the whole exercise fits in an hour if your records are in order. Here is the template.
Why auditors keep asking for this
Every major food and product-safety scheme treats the mock recall exercise as the proof that a traceability system works, not just exists. BRCGS and SQF audits expect documented traceability tests with results and timings; USDA organic certifiers commonly walk through a trace-back during inspection; and the FDA’s traceability rules are built around producing shipment records fast — a topic we cover in the FSMA 204 deadline guide. The common thread is speed under pressure. An auditor does not just ask whether you can trace a lot; they ask how long it took, because in a real recall every hour of delay is more affected product consumed.
The good news: a recall traceability test needs no consultant and no downtime. One person, one lot, one timer.
The one-hour mock recall template
Run the six steps in order and write down the clock time as you complete each one. The timings are the deliverable.
Step 1: Pick the lot (5 minutes)
Choose a lot you shipped two to six weeks ago. Newer than that and the orders are trivially fresh in memory; older and you drift out of the window where a real supplier notice would land. Do not cherry-pick your best-documented product — a randomly chosen lot is the honest test. Note the internal lot code and the supplier lot it links to; if those two are not linked, you have already found your first gap (see our lot number format guide for why both must be kept).
Step 2: Trace back (10 minutes)
Reconstruct where the lot came from: the supplier, the receipt date, the quantity received, and the certificate of analysis or other quality document for that batch. If the COA lives in an email thread rather than attached to the lot record, log it as a gap. The target is a complete upstream picture from records alone, without calling anyone.
Step 3: Trace forward (15 minutes, target: minutes not hours)
This is the heart of the exercise: produce every order that contained the lot, with customer, quantity, and ship date. In a real recall this list becomes your notification list, so completeness is everything — one missed order is one uninformed customer. This step is where spreadsheet-based operations lose entire days, and where a per-lot ledger with order references reduces the answer to a single one-click recall export. If trace-forward takes longer than the rest of the exercise combined, that is your finding.
Step 4: Reconcile the mass balance (15 minutes)
Now make the numbers close: quantity received should equal quantity shipped plus quantity on hand plus documented adjustments (damage, samples, transfers, corrections). Count the physical stock for that lot and compare. A gap of more than roughly 2 to 3 percent is a finding an auditor would write up, because unaccounted units in a real recall are units you cannot rule out having shipped. This is exactly the arithmetic an audit-ready batch ledger performs continuously, since every deduction, correction, and transfer is a ledger entry.
Step 5: Draft the customer notice (10 minutes)
Write the email you would actually send: product name, lot code, the dates affected orders shipped, what the customer should do, and how to reach you. Drafting it during a drill, when nothing is wrong, produces a calm template you can reuse under real pressure. Save it with the exercise records.
Step 6: Log gaps and fix dates (5 minutes)
List every gap the drill exposed — missing COA attachments, unlinked supplier lots, mass-balance variance, slow trace-forward — and assign each one an owner and a fix date. The log plus your step timings is the document you hand an auditor as evidence the exercise happened and led somewhere.
Scoring: how fast is fast enough?
| Result | Total time to trace (back + forward) | What it means |
|---|---|---|
| Excellent | Under 30 minutes | Recall-ready. Records are systematized; a real event is an execution problem, not a data problem. |
| Acceptable | 30 minutes to 4 hours | Workable but stressful in a real event. Fix the slowest step before the next drill. |
| Needs work | 4 hours to 2 days | A real recall would outrun your records. Prioritize a per-lot ledger over process tweaks. |
| Failing | Over 2 days or incomplete | You could not credibly notify all affected customers. Treat as an urgent project, not a finding. |
A FAILED MOCK RECALL IS A SUCCESSFUL MOCK RECALL
The exercise exists to surface gaps while they are free to fix. Finding a 6 percent mass-balance hole in a drill costs you an afternoon; finding it during a real recall costs you every customer you could not rule out.
Make step 3 a one-click step
TraceLot’s per-lot ledger records every order deduction with its order reference as you fulfill through Veeqo or Shopify, so the recall CSV — every shipped order, customer, quantity, and date for a lot — exports in one click.
See recall readiness in actionHow often to run it
Minimum cadence is once a year for any seller with dated or regulated stock; food businesses under BRCGS or SQF style schemes should run it at least twice a year, and many certifiers expect that rhythm. Beyond frequency, vary the drill so it keeps teaching.
- Alternate direction: run one drill supplier-to-customer (trace forward) and the next customer-to-supplier (trace back from a single order).
- Rotate the product and the person running the exercise, so the capability is not one employee’s memory.
- Re-test previous gaps first: the fastest audit credibility comes from showing a gap logged in one drill and closed by the next.
- Keep every exercise record — lot chosen, timings, mass balance, notice draft, gap log — in one folder per drill.
Frequently asked questions
What is a mock recall?
A mock recall is a timed drill in which a business pretends one lot of product is affected by a defect and tests whether its records can identify where the lot came from, every customer who received it, and whether received quantities reconcile with shipped and on-hand stock. No product is actually recalled.
How long should a mock recall take?
The exercise itself should fit in about an hour. Within it, the combined trace-back and trace-forward should take under 30 minutes for a recall-ready operation; more than four hours indicates records that would not keep up with a real recall.
How often should a mock recall be performed?
At least once a year for any seller handling dated or regulated products, and at least twice a year for food businesses, which matches the expectations of certification schemes such as BRCGS and SQF. Check your specific certifier’s requirements.
What is mass balance in a mock recall?
Mass balance is the reconciliation check that quantity received for a lot equals quantity shipped plus quantity still on hand plus documented adjustments such as damage or samples. A variance above roughly 2 to 3 percent is treated as a finding, because unaccounted units cannot be ruled out of a real recall.
Never ship expired stock again.
TraceLot adds batch tracking, FEFO allocation, and audit-ready records to Veeqo and Shopify. First month free.