A long clean record creates overconfidence
In Phil’s view, companies that have never experienced an outbreak are most at risk of becoming complacent, mainly because they have nothing to measure their own risk against. The lack of failure starts to feel like evidence of safety, and those are not the same thing.
Frank encountered this directly during a supplier evaluation. A company came in and led with its 80-year clean record as a credential. His immediate question was whether that record reflected rigorous searching or its absence, and he had no way to tell.
“If you've never experienced a food safety outbreak, then I think you can't help but be complacent.” — Dr. Phil Bremer
The very success that builds organizational confidence ends up eroding the conditions that made that success possible. By the time that dynamic becomes visible, something has usually already gone wrong.
Staff turnover erases institutional memory
Phil names high staff turnover as one of the clearest drivers of complacency, and specifically the departure of senior staff who aren't replaced by someone with similar depth. Written procedures can outlast every departure, but the context that gave rise to said procedure is harder to hold on to.
“Anyone who's had a food safety outbreak is very aware of the enormously emotional and economic harm it has on the company and often on themselves.” — Dr. Phil Bremer
A team that has never been through an outbreak has no equivalent reference point. A step can feel optional to someone who never learned why it exists — short-staffed or behind schedule, that's the first one to get skipped.
When companies stop seeing risk clearly
Phil calls another common failure mode the "castle on the hill" — a company with a strong track record sits behind its own walls, restricts the flow of information in and out, and gradually stops seeing what's happening beyond them.
“It’s very important that food safety professionals talk to each other, read the literature, look at recalls, and try to understand what their competitors are doing.” — Dr. Phil Bremer
Frank reads prevalence surveys from outside his own organization for exactly that reason — to stay aware of risks his own operation might not yet be catching.
How hazard analyses miss risks that are already documented
FDA regulators were inside WanaBana the year before the lead-in-cinnamon outbreak. A third-party auditing firm was there, too. Neither flagged lead in cinnamon as a reasonably foreseeable hazard, and neither did the company itself. The product was a cinnamon-flavored applesauce pouch designed and marketed for children.
“Stay up to date. The idea that you might have a heavy metal like lead in cinnamon is something to be expected.” — Frank Yiannas
The published research connecting heavy metals to spices already existed at the time of those inspections. Three separate parties had the opportunity to identify that hazard before it reached children, and the failure was the same in every case — a hazard analysis built around what had gone wrong historically rather than what the current literature said could go wrong next.
AI can improve our food safety vigilance
Companies are already using AI to surface patterns across data sets that no analyst would think to connect manually — like linking a temperature reading that's off by a tenth of a degree to three plant floor team members being out on the same shift. Individually, those details don’t necessarily trigger a flag. In combination, they might.
“We're putting together bits of the puzzle that previously we wouldn't have been able to see — within 30 seconds.” — Dr. Phil Bremer
As food companies move toward more automated systems, Frank asserts that the controls embedded in those systems will determine how risk is managed on plant floors for years to come. He warns that if food safety professionals aren't in the room when these new automated systems are designed, the controls built will reflect whoever is — and engineers and equipment vendors optimizing for throughput don't always ask the same questions about risk.
This episode of Smarter Food Safety is available now, wherever you get your podcasts.
To learn more about the host, Frank Yiannas, and why Ecolab is partnering on this show, read our profile here.
Episode Field Notes
Terms and resources worth bookmarking for food and beverage operators this year.
New Zealand Food Safety Science and Research Centre: Phil's research center, an academia-industry-regulator partnership Frank highlights as a model worth replicating: https://www.nzfssrc.org.nz/
Food Safety Culture Lab: The University of Otago lab Phil co-founded to study food safety culture as an empirical discipline, and the home base for the joint complacency study referenced in this episode: https://www.foodsafetyculturelab.com
Hazard analysis and risk-based preventive controls: The FSMA rule underlying the "hazard analysis built around past incidents" failure mode Frank and Phil discuss throughout the episode: https://www.fda.gov/food/food-safety-modernization-act-fsma/fsma-final-rule-preventive-controls-human-food
Clostridium botulinum in powdered infant formula: The hazard Frank names in this episode as one that should have been reasonably foreseeable — FDA's outbreak investigation and post-outbreak response, following 48 infant illnesses across 17 states linked to ByHeart formula in late 2025: https://www.fda.gov/food/outbreaks-foodborne-illness/post-outbreak-response-activities-clostridium-botulinum-illnesses-associated-consumption-powdered
