Environmental monitoring programs that find Listeria before the FDA does.

My CFIA-licensed (Canadian Food Inspection Agency) facility in Brantford, Ontario makes shelf-stable mushroom kits, not deli meat or sliced cheese. When I built the Preventive Control Plan I had to defend a written decision to NOT require pathogen environmental monitoring — no exposed ready-to-eat step, no in-package risk that needed verification swabbing. To make that call honestly I had to understand exactly what an environmental monitoring program should look like if I did need one. This article is that playbook. If you make food that is exposed to the air after the kill step and packed without an in-pack lethality, this is the version of the program I would hand you. I have never personally responded to a presumptive Zone 1 positive in production. The 12-step protocol in section 06 is the consensus from FDA's 2017 draft guidance, FSIS's 2014 compliance guideline, and the public records from the Bil Mar, Maple Leaf, Blue Bell, and Boar's Head outbreaks. I will say so plainly in the section.

01The honest contradiction

A program that never finds Listeria is broken.

An environmental monitoring program (EMP) is the only prerequisite program I know of where finding nothing is bad news. If you run 25 swabs a week for 18 months and never get a Listeria spp. hit, the answer is almost never "we are incredibly clean." The honest answer is usually "we are sampling in the wrong places." That is not me being cynical — that is what the FDA's January 2017 draft guidance says, and what John Butts at Land O'Frost called seek and destroy when he formalized this approach in the meat industry decades ago.

The point of an EMP is to find the spots in your plant where a pathogen is trying to set up a permanent address — what the agency literature calls a harborage site or a niche — and destroy that address before it ships product. A working EMP at a small ready-to-eat (RTE) plant typically runs 2 to 5 percent total positives, with a few percent of drains coming up positive every quarter. Zero percent across a year is a sampling deficiency, not a victory. That distinction is the entire reason this article exists.

02The trigger

When EM stops being optional.

The federal trigger for non-meat, non-poultry RTE food is 21 CFR section 117.165(a)(3) — the verification clause inside the Preventive Controls for Human Food rule. The exact words are short and worth knowing: "Environmental monitoring, for an environmental pathogen or for an appropriate indicator organism, if contamination of a ready-to-eat food with an environmental pathogen is a hazard requiring a preventive control, by collecting and testing environmental samples."

That clause is paired with 21 CFR section 117.130(c)(1)(ii) — the hazard analysis trigger. The hazard analysis "must include an evaluation of environmental pathogens whenever a ready-to-eat food is exposed to the environment prior to packaging and the packaged food does not receive a treatment or otherwise include a control measure (such as a formulation lethal to the pathogen) that would significantly minimize the pathogen."

Plain English: RTE food, plus exposure to the room after the kill step, plus no in-package lethality, equals a mandatory Listeria hazard evaluation. If that evaluation lands on "yes, this is a hazard requiring a preventive control," then EM is the verification activity you owe. There is no size exemption. A 3-person sandwich kitchen that makes RTE wraps and sells across state lines under FSMA owes the same evaluation as a 300-person sliced-deli plant.

21 CFR section 117.3 names the environmental pathogens explicitly: Listeria monocytogenes and Salmonella species. Listeria for moist RTE — deli, dairy, cut produce, sandwiches, smoked fish, hummus, soft cheese. Salmonella for low-moisture RTE — peanut butter, dry spices, dry mixes, dry pet food, infant formula, dry dairy, granola, dried fruit, cocoa, nut butters. Same Zone 1 to 4 framework either way, with the differences I cover in section 07.

For USDA-regulated meat and poultry, the framework lives at 9 CFR Part 430 with FSIS's zero-tolerance Lm policy in place since 2003. FSIS classifies RTE meat plants into Alternative 1, 2, or 3 based on whether they use a post-lethality treatment and an antimicrobial agent. EM frequency scales with the Alternative chosen. The agency's 2014 Compliance Guideline (Compliance Program 7303.040 territory) is the working document.

In Canada, the trigger is constructed through SFCR sections 47 to 89 — the Preventive Control Plan provisions. There is no single SFCR clause that says "you shall sample drains weekly." Instead the SFCR requires you to identify biological hazards and put controls in place commensurate with risk. CFIA's Listeria policy for RTE meat (aligned with FSIS for export equivalency) is what makes EM functionally mandatory for any operator shipping RTE meat into the US market.

03The zones

The 4-tier zone model — how Listeria moves through a plant.

The FDA's 2017 draft guidance — roughly 100 pages, still labeled "draft" as of this writing but cited routinely by inspectors — codifies a Zone 1 to 4 classification of every site in the plant. Risk goes down as the number goes up. Contamination flows the other way: in over time, from Zone 4 toward Zone 1, if you let it.

1. 01

   Zone 1 — Food Contact Surfaces

   Anything touching RTE food after the kill step. Slicer blades. Conveyor belts that carry product. Filler nozzles. Brining tanks. Hopper interiors. Knives. Gloves on a hand that handles exposed product. Product-contact scales. The product-contact components of a packaging machine. A Zone 1 positive is presumed product contamination until you prove otherwise.

2. 02

   Zone 2 — Non-Food-Contact, Within 3 Feet

   Equipment exteriors. Control panels. Framework legs. Conveyor undersides. Switches. Support structures. Non-contact scales. Cleaning tools staged on the line. The reason Zone 2 matters: a splash, an aerosol from a high-pressure hose, or a wipe with a contaminated cloth moves a pathogen from Zone 2 to Zone 1 in seconds. A Zone 2 positive is a red flag — one event away from product contamination.

3. 03

   Zone 3 — Production Area, Not Adjacent

   Production-room floors. Drains and drain covers. Walls. Ceilings. Forklifts. Racks. Carts. Doorways. The interior of the RTE-room HVAC. Hand-wash sinks. Employee traffic paths. Mop heads. Drains are the canary — they go positive first. A Zone 3 drain positive every quarter is normal. A Zone 3 drain positive every week at the same drain is a harborage candidate.

4. 04

   Zone 4 — Inside Facility, Outside RTE Room

   Warehouse. Raw material storage. Locker rooms. Break rooms. Hallways. Maintenance shops. Dock areas. A Zone 4 positive is informational — it tells you the building has Listeria somewhere, which is true of essentially every food building. The risk is what happens at the door between Zone 4 and Zone 3 — foot traffic, forklift wheels, gowning discipline.

Contamination flow under a working EMP looks like this: Zone 4 has positives all the time and nobody panics. Zone 3 has periodic positives in the worst spots (drains, floor-wall junctions, condensate pans) and each one gets a vector swab and a root-cause look. Zone 2 has a positive maybe once or twice a year, and each one triggers an intensified response. Zone 1 has positives rarely — and when one happens, the line stops and the 12-step protocol in section 06 fires.

If you ever see a Zone 1 positive without a corresponding Zone 2 or Zone 3 positive in the same area in the prior few weeks, your sampling is missing the harborage. The pathogen did not teleport onto the slicer blade. It came from a niche you are not sampling.

04The swab map

Building the swab map.

A swab map is the document that lists every sampling site in the plant, what zone it is, how often it gets sampled, and which week of the rotation it sits in. The 2017 draft guidance does not give you a number — it requires "adequate" relative to risk. Here is the range I would use as a starting point for a small RTE plant, validated against published ranges from Eurofins, AIB, Food Safety Magazine, FSNS, and Charm.

Frequency by zone, working from the FDA draft guidance and the meat-industry consensus:

• Zone 1. Weekly during operation for any RTE that supports Lm growth (most refrigerated RTE). Every two weeks may be defensible for RTE that does not support growth (low water activity, low pH, specific formulation). Document the reasoning in the EMP itself.
• Zone 2. Weekly to biweekly. Many high-risk plants do every Zone 2 site weekly because the splash distance from Zone 2 to Zone 1 is so short.
• Zone 3. Monthly for general sites. Drains weekly — they go positive first, they are cheap to swab, and they are the highest-information site in the plant.
• Zone 4. Quarterly for general sites, with a few sentinel sites (dock floors, locker entrances) sampled monthly to catch incoming contamination.

“

We had an EMP that pulled 12 sites a week, never found anything for two years, and I knew in my gut it was fake. When we redid the swab map with a real harborage list — drain interiors, hollow roller ends, floor-wall junctions, conveyor undersides — we found our first Listeria innocua positive in the third week. That was when the program started working.

”

Composite — small-plant QA manager

Where Listeria actually lives — the harborage list.

The consensus list from FDA, FSIS, Eurofins, Charm, AIB, AFFI, and the Cornell Food Safety Lab. If your current swab map does not include most of these sites, the map is the problem.

• Floor drains and drain covers — number one site, wet and cool and biofilm-friendly
• Floor-wall junctions and floor cracks
• Hollow conveyor rollers (unsealed interiors are a niche)
• Wheels and casters on rolling equipment
• Conveyor belt seams and undersides
• Drip pans and condensate trays
• Refrigeration coil pans
• Hollow equipment framework — legs and tubes with unsealed welds
• Slicer deflector plates and carriages
• Cleaning tools left wet between uses
• Forklift wheels traveling from cooler into production
• Cooler door gaskets and seals
• Non-filtered compressed air outlets
• Knife handles and tongs
• Hand-wash sink drains (a hand-wash sink draining into a Zone 3 floor drain is a vector path)
• Cooling unit fans and ductwork — the Bil Mar 1998 source
• HVAC drip pans and condensate lines
• Wet insulation behind walls
• Cracked welds on hollow stainless tubing
• Vacuum-sealer and thermoformer cavity interiors

Sponge beats cotton. Every time.

A study published in Food Safety Magazine in 2023 compared cotton-tip swabs against pre-moistened cellulose sponges in identical Listeria-positive environments. Cotton swabs found zero Listeria spp. positives. Sponges pre-moistened in neutralizing broth (Letheen or Dey-Engley) found Listeria spp. in 30 percent of the same sites and L. monocytogenes in 17 percent. Cotton physically traps the cells inside the fiber and cannot release them into the enrichment broth. Sponges release.

Use a cellulose or polyurethane sponge for environmental monitoring. The industry workhorse is the 3M Sponge-Stick. Other compliant options include World Bioproducts EZ Reach, Hardy Diagnostics SpongeSicle, and Hygiena Polywipe (good for large surface areas). I am not endorsing a vendor — they all work. The point is the form factor, not the brand.

Sample area for a typical Zone 1 swab is about 10 centimeters by 10 centimeters (roughly 4 inches by 4 inches). For irregular surfaces like a slicer blade or a hollow roller end, swab the whole accessible surface and document what you did. Technique: 10 vertical strokes, 10 horizontal, 10 diagonal, rotating the sponge face between passes so a fresh surface contacts the site each time.

Cost reality.

Lab pricing as of 2026 runs roughly $12 to $25 per sample for Listeria spp. with species confirmation, and $30 to $60 for whole-genome sequencing (WGS) on a Lm-positive sample to compare against PulseNet (the CDC's national outbreak database). A small plant running 25 swabs a week at $18 per sample is at $1,900 to $2,400 per month in lab fees alone. All in — labor to collect, lab fees, supplies, time to trend — a real small-plant EMP costs $3,500 to $6,000 per month. That is the honest number. A program that costs $400 a month is not a program — it is a paperwork exercise.

05Listeria spp. vs Lm

Why FDA wants you to test for the whole genus.

The FDA's 2017 draft guidance recommends testing for Listeria spp. (the whole genus) rather than only L. monocytogenes. New EMP managers find this surprising. The reasons are operational and worth understanding.

First, any Listeria species positive signals a site capable of harboring L. monocytogenes. The non-pathogenic species — L. innocua, L. welshimeri, L. seeligeri — live in the same cold, wet, organic-load niches that Lm prefers. If L. innocua can colonize a drain, so can Lm.

Second, non-monocytogenes Listeria are more common in food-plant environments. They act as a sensitive surrogate. A program testing only for Lm directly will miss niches that are about to become Lm-positive but currently host the cousin species. By the time direct-Lm testing catches up, the niche has been there for months.

The workflow is: sponge to enrichment broth, screen for Listeria spp. on selective media or by PCR, and have the lab speciate any positive. A non-monocytogenes positive still demands a corrective action — vector swab, root cause, engineering fix — because the niche is real. The pathogen just is not there yet.

There is one defensible exception. RTE meat and poultry plants operating under FSIS Alternative 1 with full post-lethality treatment plus antimicrobial agent sometimes test for Lm directly because the regulatory framework is built around Lm specifically. Most other RTE plants — dairy, produce, sandwich, smoked fish, bakery — should follow the FDA recommendation and cast the spp. net.

06The positive Zone 1 protocol

What to do when Zone 1 comes back positive.

This is the section you cannot write under stress. It has to be written, signed, and mock-rehearsed before the call comes in at 6 AM on a Tuesday. Below is the consensus 12-step protocol from the FDA 2017 draft guidance, the FSIS 2014 compliance guideline, and the public records of the Bil Mar, Maple Leaf, Blue Bell, and Boar's Head responses. I have not personally executed this protocol in production. My facility does not have a Zone 1 trigger. This is the version I would hand to a small RTE plant building their own.

1. 01

   HOLD the product

   All product produced on the implicated line or equipment since the last Zone 1 negative goes on hold. Worst case lookback: last positive-to-positive period, or the last 30 days, whichever is longer. Mark it, log it, lock it. Do not ship.

2. 02

   Cease production on the implicated line

   Switch to a different line if you have one. Shut down if you do not. Producing more product through a known-contaminated line buys you nothing and adds product you will eventually have to destroy.

3. 03

   Notify management and document the trigger

   Time of result. Method (sponge, swab area, sample technician). Site code. Sponge lot. Sample ID at the lab. The minute you receive the call from the lab is the minute the inspector's clock starts in the event you ever get there. Document everything.

4. 04

   Confirmatory testing — speciate the result

   Listeria spp. positive does not automatically mean L. monocytogenes. Run the confirmation — 48 to 72 hours for traditional culture, 24 hours for rapid methods (BAX, VIDAS, MALDI-TOF). Do not move forward without it. Equally important — do not release product based on the absence of confirmation. Hold until you know.

5. 05

   Vector swab — 4 to 10 samples within 3 feet, same day

   Star pattern around the original positive site. The point of vector swabbing is to answer one question: is this a point source (one contaminated spot) or is it spreading (multiple positives in the star)? Vector results in 24 to 48 hours steer everything else.

6. 06

   Intensified cleaning and sanitation (ICS)

   Full disassembly of the implicated equipment. Stronger chemistry — peracetic acid at 80 to 200 parts per million is a common upgrade from the routine quat or chlorine. Photo-document every step. ICS is not "clean it harder" — it is a documented escalation that the inspector will look for.

7. 07

   Verification swab — 24 hours after ICS

   Re-swab the original site and every vector site. Three consecutive negative rounds is the minimum to return to normal operation. FDA prefers more than three when the original was Lm. The pattern matters — three consecutive negatives over three weeks is far stronger than three swabs taken back to back the same day.

8. 08

   Product disposition decision

   Confirmed L. monocytogenes on a Zone 1 site plus RTE product that supports Lm growth is, in most realistic scenarios, a recall. Destruction or rework with a validated kill step (rare for RTE) are the alternatives. This is the decision that costs millions and ends careers — write the decision tree before the moment arrives so the call is consistent.

9. 09

   Root cause analysis — real, not theater

   A real 5-Why or fishbone analysis, not a one-line "employee training" finding. The Boar's Head and Blue Bell records both show what theatrical RCA looks like: the same root cause cited month after month, with no engineering change between citations.

10. 10

    Corrective action — engineering change if design-related

    Hollow roller with unsealed welds — replace with a sealed model. Cracked floor coating — repour. Drain that goes positive every quarter — disassemble, replace traps and grates, possibly demolish and rebuild the floor section around it. A cleaning frequency change without an engineering change usually fails within months.

11. 11

    Notification

    FDA or FSIS as required. Customers under contract terms. Your insurance carrier. Your legal counsel. The order matters and your written EMP should specify who calls whom in what sequence.

12. 12

    Extended verification — 4 to 8 weeks of intensified swabbing

    The point is to confirm the niche is actually destroyed, not just dormant. Three negatives the week after ICS is not enough. Sustained negatives over a month or more, while production runs at normal volume, is the standard FDA wants.

07Routine vs investigative

Two kinds of swabs, two different jobs.

The routine swab map and the investigative response use the same physical technique but answer different questions. Mixing them up is one of the most common failure modes in small-plant programs.

The other rule that comes from this distinction: never "clean and re-swab the same spot." A single re-swab of an original positive site is not corrective action. It is hope. Vector swabbing — 4 to 10 samples in a star — is the only honest way to know whether the niche is point or spreading.

08Trending

The difference between a real EMP and a checkbox EMP.

The most common deficiency in small-plant EMPs is no trending. A program that samples 25 sites a week and files the negatives without analysis will miss a harborage establishing in real time. Here is the minimum trending framework.

1. 01

   Positive percentage (PP)

   Total positives divided by total samples, calculated monthly and rolling 90 days. A working high-risk RTE program runs 2 to 5 percent Listeria spp. overall, with 0 to 1 percent on Zone 1, 5 to 10 percent on Zone 3 drains. Zero percent across 90 plus days means your sampling is wrong, not that your plant is clean. Sustained 8 percent or higher on Zone 2 means you are losing control.

2. 02

   Site-level trending

   Flag any site that comes up positive more than 3 times in a rolling 90-day window. That is the operational definition of a candidate harborage site. Standard response is a deep investigation — disassemble the equipment, look behind walls, inspect every weld, replace gaskets, consider retiring the equipment.

3. 03

   Vector swab on every positive

   Already covered in section 06 and section 07. Every positive gets a vector. No exceptions. If you skip the vector you lose the only data that distinguishes point source from spread.

4. 04

   Speciation tracking

   Track L. innocua separately from L. monocytogenes. An innocua trend at a site is predictive — the niche is there, the cousin is already in it, Lm is the next visitor. Investigate as if Lm.

5. 05

   PFGE or WGS strain tracking for repeat positives

   Three Zone 3 drain positives in 90 days that share the same whole-genome sequence pattern is a resident strain. Every major listeriosis outbreak in the historical record (Bil Mar 1998, Maple Leaf 2008, Blue Bell 2015, Boar's Head 2024) traced to a resident strain that had been in the building for months or years. Detecting resident strains is exactly what the EMP exists to do.

09Low-moisture and Salmonella

Salmonella EMPs for low-moisture foods.

For low-moisture RTE — peanut butter, dry spices, flour, dry mixes, cereal, dry dairy powders, infant formula, dried fruit, nut butters, chocolate, dry pet food, supplements — the environmental pathogen is Salmonella, not Listeria. Same Zone 1 to 4 framework. Several important differences in the practice.

• Salmonella thrives in dry niches that Listeria ignores. Sifter bags, dust collectors, dryer interiors, scoops, totes, bucket elevators, conveyor housings, the seams of pneumatic-conveying systems.
• Wet cleaning is discouraged in dry plants. Introducing water creates new wet niches and can reactivate dormant Salmonella cells. Dry cleaning (vacuum, sweep) with targeted spot wet-clean is the norm. This is the opposite of the Listeria playbook.
• Action limits are tighter. A single Salmonella positive on a Zone 1 surface is typically a recall trigger, not a "hold and verify" event.
• Useful indicator organisms include Enterobacteriaceae, coliforms, and total aerobic plate count. They serve a similar role to Listeria innocua in the wet-plant world — present in the same niches, easier to detect, predictive of pathogen risk.
• Reference documents: the American Spice Trade Association (ASTA) "Comprehensive Guidelines for Salmonella Control" and FDA's draft guidance on Salmonella in low-moisture foods (in development as of writing).

The 2009 Peanut Corporation of America (PCA) Salmonella outbreak — 714 illnesses, 9 deaths, and the only food-safety case where a CEO was sentenced to 28 years in federal prison — is the low-moisture equivalent of the Bil Mar and Boar's Head benchmark. PCA's environmental record showed positives that were either ignored or unreported.

10Scheme requirements

What SQF, BRCGS, and FSSC actually require.

If you carry a Global Food Safety Initiative (GFSI) certification on top of your regulatory baseline, the audit framework adds its own EM expectations. The short version of the current editions:

• SQF Edition 10 (effective May 24, 2024). EM is mandatory for sites manufacturing high-risk RTE products with pathogen exposure. The requirements are risk-based: documented sampling plan with pathogen or indicator swabbing, sample identification, frequency, action limits, trending, and corrective action. SQFI Tip Sheet 19 on environmental monitoring is the working guidance. EM is a Critical clause for the relevant product categories — a failure on EM at audit will not be a minor finding.
• BRCGS Issue 9 clause 4.11.8 (effective February 1, 2023). Where the risk assessment indicates an EMP is required, the program must include a documented sampling plan, locations and frequency, the target pathogens or indicator organisms, test methods, recording of results, action limits and corrective action protocols, and a periodic review of the programme. Issue 9 tightened the trending and action-limit expectations compared to Issue 8 — auditors now want to see the trend chart, not just the spreadsheet of results.
• FSSC 22000 v6. Inherits the ISO 22002-1 prerequisite-program requirements plus the FSSC Additional Requirements, which include explicit environmental monitoring expectations for categories C (perishable animal products), I (transport and storage), K (catering), and several others.

The CFIA framework for SFCR-licensed operators does not have a single "EM clause" the way FDA does. Instead the requirement is constructed from the PCP biological-hazard analysis under SFCR sections 47 to 89 plus CFIA's Listeria policy for RTE meat. Operators exporting RTE meat into the US under the FSIS equivalency arrangement are reviewed against FSIS expectations, which makes the FDA and FSIS programs the working reference.

11Outbreaks

The four outbreaks that built this rule.

Every paragraph of the FDA's draft guidance has a body count behind it. The named cases are not history lessons. They are the operating manual the agency wrote in the years afterward. Restrained language here because real people died and the harm to families is permanent.

1. 01

   Bil Mar Foods / Sara Lee — Zeeland, Michigan, 1998-1999

   Demolition of an aging refrigeration unit in July 1998 introduced contamination into the production area. Public records and the subsequent CDC investigation indicate plant management had instructed lab technicians to test only for conditions favorable to Listeria, not for the organism itself. 100-plus illnesses across 22 states, 15 deaths, 6 miscarriages, 35 million pounds of product recalled. Sara Lee pleaded guilty to misdemeanor charges and paid a $200,000 fine plus a $4.4 million research grant. CDC MMWR 49(50);1129-1130.

2. 02

   Maple Leaf Foods — Toronto, 2008

   Slicer harborage on Lines 8 and 9 at the Bartor Road facility. 57 cases, 23 deaths. $27 million class-action settlement. The Maple Leaf case is the catalyst for modern CFIA Listeria policy and for the SFCR PCP framework. CEO Michael McCain's crisis response is studied in business schools, but the EMP lesson is simpler — Lines 8 and 9 had detectable Listeria signals that the program of the day did not act on.

3. 03

   Blue Bell Creameries — multiple plants, 2010-2015

   Operators still wrongly believe ice cream cannot support Listeria. The mistake comes from confusing the frozen finished product with the production environment. Exposure happens before the freeze. Blue Bell's Broken Arrow, Oklahoma plant recorded 17 environmental Listeria positives between March 2013 and February 2015 with inadequate investigation and no product holds. 10 cases, 4 states, 3 deaths. FDA Form 483s issued in 2015 cited failure to investigate environmental positives. Blue Bell pleaded guilty in 2020. The 17-positives number is the canonical example of what "found and ignored" looks like.

4. 04

   Boar's Head — Jarratt, Virginia, 2024

   61 cases, 19 states, 60 hospitalizations, 10 deaths. The largest US listeriosis outbreak since 2011. 7 million pounds of product recalled. USDA-FSIS records show 69 noncompliances at the Jarratt plant between August 2023 and August 2024 — dripping condensate on exposed product, fans blowing condensate onto exposed product, "ants traveling down the wall," cockroach sightings, "heavy meat buildup on room walls," black mold and mildew, "flies entering vats of pickle." FSIS suspended production on July 31, 2024. Boar's Head closed the plant indefinitely in September 2024 and announced a permanent halt to liverwurst production. Federal inspectors took over inspection authority in 2025.

The anatomy is identical across all four cases: a niche, a missed signal, an inadequate response. The EMP exists to break that chain at the second step.

12The inspector

What "adequate" looks like to the inspector.

FDA and FSIS inspectors looking at an EMP look for five specific things. Knowing the list lets you self-audit before they show up.

1. 01

   Risk basis

   The program is informed by your actual hazard analysis (the PCP or the FSP), not by a copied template. Different products, different processes, different exposure points, different EMPs. If your EMP looks exactly like the one in a textbook, the inspector will ask how the hazards in your plant drove the design.

2. 02

   Coverage

   The sites a knowledgeable microbiologist would expect to be hot. Experienced inspectors walk to the drain first. If your swab map does not include drains, drain covers, hollow roller ends, conveyor undersides, floor-wall junctions, and condensate pans, the map is not adequate.

3. 03

   Frequency

   Often enough to detect periodic harborage establishment. Monthly Zone 1 sampling on a high-Lm-risk RTE product is hard to defend. Weekly is the working standard.

4. 04

   Response history

   When positives were found — and they will have been, if the program is real — did the facility respond proportionately? Did the corrective action destroy the niche, or just clean it? "Cleaned and re-swabbed, no further action" is the language that gets cited.

5. 05

   Trending

   Is somebody actually looking at the data over time? A binder of results filed sequentially with no aggregation is filed evidence, not trending. The inspector wants to see the dashboard, the rolling 90-day site list, the strain tracker.

A program with zero Zone 3 positives in 18 months is suspect on its face. A program with positives found and not acted on is damning. The pattern that satisfies an inspector is: found, vector-swabbed, root cause identified, engineering change made, three negative verification rounds documented, extended verification over weeks confirming the niche is destroyed.

13Templates

Start with the logs that touch sanitation.

If you are building toward a full EMP, you are also rebuilding the daily sanitation discipline the EMP verifies. The cleaning logs below are the working paper version. Use them on a tablet or print them to a clipboard. No account, no email required.

The cleaning logs are the upstream record that makes EM results interpretable. A Zone 3 positive at a drain that was logged as cleaned the same morning tells a different story than a Zone 3 positive at a drain nobody touched for a week. The EMP without the sanitation log is half a system.

14The handoff

When you are ready to put the swab map in the same system as the plan.

The EMP belongs in the same record system as the HACCP plan or PCP it verifies. Trying to run it in a separate spreadsheet that nobody opens is exactly how Boar's Head and Blue Bell went year after year with positives that lived in one binder and corrective actions that lived in another.

Footnotes

1.21 CFR §117.165 — Verification (eCFR) — ecfr.gov

2.21 CFR §117.130 — Hazard analysis (eCFR) — ecfr.gov

3.FDA Draft Guidance: Control of Listeria monocytogenes in Ready-To-Eat Foods (January 2017) — fda.gov

4.USDA-FSIS Compliance Guideline: Controlling Listeria monocytogenes in Post-Lethality Exposed RTE Meat and Poultry (2014) — fsis.usda.gov

5.9 CFR Part 430 — RTE meat and poultry Lm — ecfr.gov

6.CDC — Boar's Head Outbreak Investigation 2024 — cdc.gov

7.CDC MMWR — 1998-1999 Listeriosis Outbreak (Bil Mar / Sara Lee), 49(50);1129-1130 — cdc.gov

8.PHAC — Lessons Learned: 2008 Listeriosis Outbreak (Maple Leaf) — canada.ca

9.Cornell expert elicitation: Lm EMP site selection (Food Control, 2017) — sciencedirect.com

10.Food Safety Magazine — Cotton vs Sponge Recovery Study (2023) — food-safety.com

Andrew Langevin·CFIA-licensed facility, Brantford ON· Published 2026-06-04· 13 min read· Wikidata Q139112497