Why a mock recall is worth practicing

The short answer. Under the FSMA 204 Food Traceability Rule, FDA can ask a firm that handles foods on its Food Traceability List for an electronic sortable spreadsheet of the relevant traceability records within 24 hours during an outbreak, recall or other threat to public health, and FDA has said it will not enforce the rule before July 20, 2028 (FDA). A mock recall tests that ability by tracing one lot back to its grower and forward through every production run it went into, every new lot those runs made, and every customer location that received any of them. The records follow three kinds of events, receiving, transformation and shipping, joined by traceability lot codes kept exactly as written.

A recall rarely starts with a clear question. It starts with a phone call: a customer, a grower or FDA asks where a particular lot went. The answer has to come together quickly, from receiving logs, production sheets and shipping records that were never kept with that question in mind.

Under the FSMA 204 Food Traceability Rule, FDA can ask a firm that handles foods on its Food Traceability List for an electronic sortable spreadsheet of the relevant records within 24 hours during an outbreak, recall or other threat to public health (FDA). FDA has said it will not enforce the rule before July 20, 2028. Many retailers and foodservice distributors are already writing traceability into supplier contracts, so the question may come sooner from a customer than from FDA.

A mock recall is how you find out, on a quiet day, whether you can answer it.

Last reviewed: November 11, 2026

What the trace has to follow

The rule organizes the records around three kinds of events, and a mock recall follows a lot through all three.

  • Receiving: the traceability lot code, quantity and unit, product description, who you got it from, where it was received, the date, the lot code's source, and the reference document, such as a bill of lading.
  • Transformation: for each production run, every input lot and how much of it was used, and the new lot made, with its own lot code, date, quantity and reference document.
  • Shipping: the lot code, quantity and unit, who it went to, where it shipped from, the date, the lot code's source and the reference document.

A trace goes two ways. Back, from the lot to the grower who sent it. Forward, from the lot to every production run it went into, every new lot those runs made, and every customer location that received any of them. The forward trace is the hard one, because each new lot starts its own branch.

A worked example

Here is an illustrative case, with every number made up.

You receive 40 cases of cilantro under the grower's lot code. Over the next three days, that lot goes into three production runs: two batches of salsa and one of a herb mix. Each run uses other lots too, and each makes a new lot under your own code. Those three new lots ship on eleven trucks to five customers, and one salsa lot is split across four of them.

A mock recall on the cilantro lot has to produce:

  1. the receiving record for the 40 cases;
  2. three transformation records, each listing the cilantro lot among its inputs and the new lot it made;
  3. eleven shipping records, one for each truckload of a new lot;
  4. a list of the five customers and their locations, the people you would call first.

That is fifteen records from three different logs, joined by lot codes that someone typed by hand. Miss one transformation record and a whole branch of shipments drops out of the answer, without anything looking wrong.

How to do it better, even in a spreadsheet

A spreadsheet can carry a mock recall if it is laid out for one.

  1. Three tabs, one row per event. Keep Receiving, Transformation and Shipping as separate tabs, with the same column names on every tab for the lot code, product, quantity, unit, date and location. FDA publishes a template with one tab per kind of event; it says the template is illustrative and you don't have to use it, but it is a sensible layout to copy.
  2. Store lot codes as text. Otherwise a spreadsheet drops leading zeros and turns some codes into dates. A lot code has to be kept exactly as the grower wrote it, character for character.
  3. One row per input lot. A production run that used two lots of cilantro gets two input rows, both pointing to the same new lot. This is where spreadsheets most often lose the trail.
  4. Never leave a lot code blank. If a delivery arrives without one, give it a temporary code and flag the row, so it is traced from the start.
  5. Trace with filters, branch by branch. Filter Transformation for the lot, note every new lot, then filter Shipping for each of those, and repeat if a new lot went into another run.
  6. Keep a drill log. Write down the lot, the start time, the end time, who ran it and what you found missing. That log is what a food safety plan or a customer audit asks to see.

The weak point is step 5. Every branch is another round of filtering by hand, and nothing tells you when you have missed one.

How a system does it

The demo for Harrow Creek Produce, a fictional fresh-cut produce processor, shows the same drill done by the system.

Receiving keeps every receiving data element, one line per lot, and a line with no lot code gets a temporary code and a flag. When a grower emails a bill of lading, the AI helper reads the PDF into a draft receipt, with lot codes character for character, and lists what it was unsure of first. A person checks the draft beside the PDF and receives the truck; nothing is received on the AI's word alone. Production runs record the lots used and the new lot made, and held lots aren't offered. Shipping is recorded by lot to each customer location.

The mock recall screen then does the branching for you. Pick a lot, start the clock, and the system shows every run, lot and customer the lot touched, back to the growers and forward to every customer. Download the sortable spreadsheet, laid out column for column like FDA's template, with its Shipping, Receiving and Transformation tabs. The drill log keeps the time from start to spreadsheet. If the drill turns up a real problem, the lot and everything made from it can be put on hold in one step.

In the demo's sample data, one 60-case romaine lot went into nine production runs and out on twenty trucks to all ten customers within ten days. The system works out that trace in a fraction of a second. The time a drill takes is your team's, reviewing what it found.

The system keeps the records and produces the spreadsheet. The written traceability plan and procedures the rule asks for are still yours.

Try it, and what it costs

The food traceability page explains the whole system. The live demo lets you sign in with one click, with no account to create, and run a mock recall to the spreadsheet yourself. The data resets every night, so you can't break anything.

A typical build costs $8,000 to $25,000 at a fixed price, and hosting and support start at $350 a month.

Questions

When will FDA enforce the FSMA 204 Food Traceability Rule?

FDA has said it will not enforce the rule before July 20, 2028. Many retailers and foodservice distributors are already writing traceability into supplier contracts, so the request may come from a customer before it comes from FDA.

Do I have to use FDA's sortable spreadsheet template?

No. FDA publishes a template with one tab per kind of event and describes it as illustrative and not required, though it is a sensible layout to copy.

What records does a mock recall have to pull together?

The receiving record for the lot, every transformation record that lists the lot as an input along with the new lot it made, every shipping record for those new lots, and the list of customers and locations that received them. Missing one transformation record drops a whole branch of shipments from the answer without anything looking wrong.