Food safetyMay 3, 202612 min read

FSMA Section 204 traceability: extended to July 2028, but the work has not changed

In November 2025 Congress directed FDA not to enforce the Food Traceability Rule before July 20, 2028. The 30-month reprieve does not change the rule itself — only the deadline. Here is what food labs should do with the extra time.

On August 7, 2025, FDA proposed a 30-month extension of the Food Traceability Final Rule's compliance date, moving it from January 20, 2026 to July 20, 2028. In November 2025, the Continuing Appropriations, Agriculture, Legislative Branch, Military Construction, and Veterans Affairs Extensions Act of 2026 directed FDA not to enforce the Food Traceability Rule prior to July 20, 2028. The extension is real. The rule itself is unchanged. FDA has stated explicitly that the requirements are not being relaxed — only the deadline is moving.

For food labs and supply-chain operators, the temptation is to put the FSMA 204 binder on a shelf for two years and pick it up in 2028. That is a mistake for three reasons we can walk through. The work the rule requires is not implementation work that scales linearly with calendar time. It is process and data-architecture work that compounds — and labs that use the extra 30 months to do it properly will be in a different operational position by July 2028 than labs that defer.

What FSMA 204 actually requires

Section 204 of the Food Safety Modernization Act creates a Food Traceability Final Rule that applies to entities — manufacturers, processors, packers, and holders — handling foods on the Food Traceability List (FTL). The list includes leafy greens, certain melons, herbs, sprouts, fresh-cut fruits and vegetables, finfish, shellfish, ready-to-eat deli salads, soft and fresh soft cheeses, shell eggs, and nut butters. It also applies to foods containing FTL items as ingredients when the listed food remains in the same form.

Entities subject to the rule must capture and maintain Key Data Elements (KDEs) at each Critical Tracking Event (CTE). The CTEs are: harvesting, cooling, initial packing, first land-based receiver (for fish), shipping, receiving, and transformation. Each CTE has its own KDE set — at minimum the location, dates, quantity, unit of measure, and traceability lot code. Records must be available to FDA within 24 hours of request.

That last requirement — 24 hours — is the operational heart of the rule. It is not enough to have the records. They must be retrievable, complete, and unambiguously linked from a finished SKU back through every CTE in its supply chain.

Why the extension does not change the workload

The rule is not relaxing

FDA was unambiguous in the extension announcement: the 30 months is to allow industry-wide coordination across thousands of entities. The rule's requirements — the FTL, the KDEs, the CTEs, the 24-hour retrieval — are not being revised. Labs and processors that thought a delay might mean a watering-down should re-read the August 2025 Federal Register notice. It does not.

The traceability lot code is the architectural problem

The KDE that breaks most existing systems is the Traceability Lot Code (TLC) and the requirement to link it across CTEs. A processor receiving leafy greens has to capture the TLC assigned by the upstream packer, link it to the new TLC assigned at transformation (when the greens are shredded and bagged), and pass that linkage forward to the next entity in the chain.

Most legacy food-safety and ERP systems track lots internally but do not track the upstream TLC — the lot the supplier assigned. Many do not record the linkage on transformation. Both gaps invalidate FSMA 204 traceability the moment FDA traces an outbreak through the chain.

This is a data-architecture problem, not a calendar problem. Adding TLC fields to a system that was not designed for them, populating them retroactively, and validating that the linkage works for an outbreak query that pulls KDEs across multiple CTEs takes months. Labs that started in 2024 are mostly through it. Labs that start in 2027 will have nine months at most.

Trading partners are still moving

Major retailers and food-service operators have informed their suppliers that FSMA 204-compliant data exchange will be required regardless of FDA's enforcement date. Walmart, Sysco, US Foods, and Whole Foods have all signaled that their internal traceability programs are tied to the original 2026 timeline, and supplier compliance with their data formats is a commercial requirement independent of FDA.

For suppliers, this means the practical compliance deadline is set by the largest customer's onboarding requirement, not by the federal enforcement date. A supplier that loses its retailer slot in mid-2026 because its EDI integration cannot pass TLCs in the agreed format does not get to argue that FDA pushed the rule to 2028.

What labs and processors should do with the 30 months

The teams that benefit most from the extension are using it for the work that would have been rushed under the original timeline. In rough order of leverage:

1. Map the actual supply chain, end to end

Most food businesses have a simplified mental model of their supply chain that does not match reality. A lab that thinks it sources from 12 growers may discover, when it pulls supplier records, that it actually sources from 47 — because three of the 12 are aggregators with their own subsuppliers. FSMA 204 traceability requires the real chain, not the simplified one.

The mapping exercise is unglamorous and time-consuming. It is also where most of the unrecorded data lives. Spending six months in 2026 walking the real chain is cheaper than spending six weeks in 2028 trying to backfill it.

2. Standardize the traceability lot code generation logic

The rule does not specify the TLC format. Industry consensus is converging on GS1-compliant Global Trade Item Numbers plus batch/lot, but many smaller operators still use internal codes that do not interoperate. Standardizing the TLC format internally is a one-time engineering task. Doing it for the first time under enforcement pressure with no time for parallel running is a reliability nightmare.

3. Validate the 24-hour retrieval workflow

The rule requires producing records within 24 hours of an FDA request. Labs should run that drill — internally — before it is real. Pick a finished SKU at random, request its full traceability record, and time how long it takes to assemble. If the answer is "a week, with three people pulling from four systems," the system is not ready.

Doing the drill quarterly throughout 2026-2027 reveals the actual gaps before they show up under regulatory pressure. Most gaps are not about missing data — they are about data scattered across systems with no single retrieval path.

Operational reality. The labs that pass simulated 24-hour retrievals have one of two architectures: a single LIMS with explicit traceability fields, or a LIMS plus a thin traceability layer that pulls from multiple systems on query. Labs that try to query four databases and stitch the result manually do not pass the simulation in 24 hours, let alone the real thing.

4. Train the receiving dock

The largest source of incomplete KDEs is the receiving dock. A pallet arrives, a paper BOL is scanned, the receiver enters quantity and date, and the upstream TLC never makes it into the system because the field was optional. Until receiving capture is structurally complete — operator cannot close the receiving event without the TLC — the rest of the traceability layer is built on incomplete data.

This is a UI-and-process problem that takes months to land properly. Operators resist new required fields. Supervisors push back when the receiving line slows. Finance pushes back when training hours show up in the budget. Working through that change management is exactly the kind of multi-quarter project the extension makes feasible.

5. Validate transformation linkage

Transformation — taking received raw materials and turning them into a different SKU — is where TLC linkage breaks most often. The rule requires recording the input TLCs, the output TLC, the date, and the location. In practice, transformation events in many systems either record only the output lot or record inputs as a free-text note that cannot be queried.

Labs running our food module get this linkage natively — every transformation event records inputs as structured TLC references rather than free-text — but the same architectural property is what to look for in any system being evaluated. If transformation linkage is a free-text field, it is not FSMA 204 traceability.

What not to do with the extension

  • Do not pause the project.The teams that paused FSMA 204 work after the extension announcement and plan to restart in 2027 will be in the same scramble that the original timeline would have caused, just shifted. The rule's implementation cost is not in the last six months — it is in the data architecture, which compounds.
  • Do not assume further extensions. FDA has been clear that the 30 months is the runway. Congressional language tied to an appropriations act is not a permanent shift. Planning past July 2028 is wishful.
  • Do not assume the FTL will shrink. The Food Traceability List has been criticized by some industry groups, and there has been speculation about narrowing it. FDA has not signaled any intent to do so. Building for the current FTL is the only defensible plan.
  • Do not let trading partners drive a different timeline. If your largest retailer has their own 2026 internal milestone, the regulatory deadline is irrelevant to your commercial relationship. Comply with the customer date, not the federal date.

The compounding advantage

FSMA 204 is, fundamentally, a data-architecture rule. Labs and processors that internalize that — and use the extension to fix the architecture rather than to defer the project — end up in a position where traceability becomes a query rather than an investigation. That capability has value beyond FDA compliance: it cuts recall scope, accelerates internal root-cause analysis, and reduces customer onboarding friction with major retailers.

Labs that defer end up doing the same architectural work in 2028 under enforcement pressure, with less time, less testing, and more risk. The extension is generous. It is not a reprieve. It is permission to do the work properly the first time.