HACCP Verification: The Part of Your Plan the Auditor Reads Most Carefully
Everyone obsesses over monitoring — the CCP logs, the critical limits, the frequencies. And monitoring matters. But here’s something experienced QA managers learn: auditors spend disproportionate time on verification. Because monitoring says “we did the checks.” Verification says “the checks actually work.”
It’s the difference between a plan that’s performed and a plan that’s proven. And the Codex General Principles of Food Hygiene — the foundation under every HACCP scheme on earth — treats verification as a principle in its own right, not an afterthought. Here’s what auditors probe.
Validation vs. verification: the distinction that trips people up. These get confused constantly, so let’s settle it. Validation happens before (or when you change things): it’s the evidence that your control measure can work — the scientific basis, the challenge study, the published data showing that this time-temperature combination kills this pathogen in this product. Verification happens ongoing: it’s the evidence that the control is working — calibration, record review, testing, audits.
The auditor’s question on validation: “Show me the scientific basis for this critical limit.” If your cook CCP is 74°C, where did that number come from? Regulatory guidance? A published thermal death time study? A challenge study on your product? “We’ve always used 74” is not validation. Neither is “the consultant set it.” You need the basis documented — and if the basis is a regulation or published standard, cite it specifically. We’ve seen HACCP plans with critical limits that were simply wrong for the product — a 74°C limit applied to a product where the actual hazard needed a different time-temperature combination — because nobody ever checked the science. The auditor did.
Record review: verification’s daily heartbeat. Someone other than the person who did the monitoring should review CCP records — daily, or at least before product ships. The auditor checks: is the review happening? Is it timely? Does the reviewer actually look (initials on every page with no evidence of reading don’t count)? And critically: what happens when the reviewer finds a problem? A review that catches a missed entry or an out-of-spec reading and triggers action is verification working. A review that’s a rubber stamp is theater.
Calibration as verification. Your CCP thermometer’s calibration records are verification evidence — they prove the monitoring data was accurate. The auditor connects these dots explicitly: monitoring record → instrument used → calibration status of that instrument. Break any link and the verification chain fails. This is why the calibration master list should identify which instruments monitor CCPs — it makes the linkage auditable instead of archaeological.
Targeted testing: the reality check. Verification often includes product or environmental testing — not as a control (testing can’t make unsafe food safe), but as a check that the system works. Finished product micro testing, environmental pathogen monitoring (especially for Listeria in ready-to-eat environments, Salmonella in dry environments) — the auditor wants to see that testing exists, that it’s risk-based, and — the part everyone forgets — that there’s a procedure for what happens on a positive. A positive with no documented response is worse than no testing at all, because it proves you knew and didn’t act.
The annual HACCP review: real or ritual? Most schemes require periodic reassessment of the HACCP plan — typically annually, plus whenever something changes. The auditor reads the review record looking for evidence of thought: Were the hazard analysis and CCPs reconsidered? Did complaints, deviations, or new science trigger changes? Or does the review say “no changes needed” every year in identical wording? A plan that never changes is either perfect (unlikely) or unexamined (likely). The best review records we’ve seen document the discussion — “considered whether the new supplier’s ingredient changed the hazard profile; concluded no change needed because…” — showing the team actually thought. That’s what reassessment means.
Verification of the verification. It gets meta, but auditors do check whether your verification activities themselves are working: Are internal audits covering HACCP? Does management review look at verification data? Is there trending of CCP deviations? A deviation trend going up is the system telling you something’s degrading — the verification program should catch it before the auditor does.
What a strong verification program looks like on paper: validation documentation for each CCP (the scientific basis, filed with the plan); a verification schedule (what gets verified, how often, by whom); calibration records linked to CCP instruments; record review records (timely, by an independent reviewer); testing plans with response procedures; the annual reassessment record with evidence of genuine review; and trending that feeds management review.
Here’s the mindset: monitoring is the doing. Verification is the proving. An auditor can watch you monitor for a day and see diligence. But diligence without proof is just effort. The verification program is what turns your HACCP plan from a set of activities into a validated, verified system — which is what the standard actually requires, and what keeps the certificate on the wall.
Pull your HACCP plan’s validation references right now. Can you put your finger on the scientific basis for each critical limit? If not, you’ve found your pre-audit homework.