Performance Monitoring in a Seveso VBS: What, Why, and How

"Continuous improvement" is a phrase that gets used loosely in safety management. On a general worksite, it's a philosophy - a commitment to get a little better over time. At a Seveso-classified establishment, it isn't a philosophy. It's a legal requirement with a name: prestatiebewaking, the performance monitoring element of your veiligheidsbeheersysteem (VBS), mandated under Annex III of the Seveso III Directive (2012/18/EU).

The distinction matters more than it looks. If you treat performance monitoring as a generic continuous-improvement exercise, you'll build something that feels productive but doesn't hold up under inspection - because a Seveso VBS isn't judged on whether you're "getting better." It's judged on whether you can show, with data, that you know how your major-hazard controls are actually performing.

This post covers what prestatiebewaking requires under Seveso III, why the standard continuous-improvement playbook undershoots that requirement, and how inspection and audit software fits into closing the gap.

What Prestatiebewaking Actually Requires

Annex III of Seveso III sets out the elements every VBS must contain. Performance monitoring is one of them, sitting alongside risk assessment, operational control, management of change, emergency planning, and audit and review. Its purpose is narrow and specific: adopt and implement procedures for the ongoing assessment of compliance with the objectives set by the operator's major-accident prevention policy (MAPP), and mechanisms for investigation and corrective action when those objectives are not met.

In practice, that breaks down into a few concrete obligations:

Track leading and lagging indicators, not just incident counts. Lagging indicators (accidents, dangerous occurrences, losses of containment) tell you a barrier already failed. Leading indicators - overdue inspections, expired certifications, backlog on corrective actions, drift in permit-to-work compliance - tell you a barrier is about to fail. A VBS that only tracks the lagging side is monitoring outcomes, not performance.
Analyze trends in incidents and near-misses, not just log them. A near-miss register that nobody analyzes for recurring root causes or common failure points isn't performance monitoring - it's recordkeeping. The requirement is to identify patterns: which equipment, which procedures, which sites or shifts generate disproportionate findings.
Verify that safety-critical equipment and procedures perform as designed. This includes the inspection and testing regime for safety-critical elements identified in your risk assessment - pressure relief systems, gas detection, fire and gas systems, containment integrity - and confirming those checks actually happen on schedule, not just that they're scheduled.
Feed findings back into the management system. Monitoring data has to close the loop: results feed corrective actions, corrective actions get tracked to completion, and the MAPP and procedures get revised when monitoring shows they're not achieving their objectives. A monitoring process that produces reports nobody acts on fails the requirement regardless of how much data it collects.
Produce evidence a competent authority can review. Under Seveso III, monitoring isn't just an internal management tool - it's something inspectors will ask to see during a Seveso inspection. That means the data needs to be retrievable, dated, attributable, and traceable to the specific MAPP objective or risk control it relates to.

Why Generic Continuous Improvement Advice Falls Short Here

Standard continuous-improvement frameworks - Kaizen, PDCA, lean-style "small changes compound" thinking - aren't wrong. They're just calibrated for a different risk profile than a major-hazard establishment operates under.

The stakes change what "good enough" means. On a general worksite, an unaddressed near-miss might mean a future minor injury. At a Seveso site, an unaddressed near-miss in a process safety context can be a precursor to a major accident - a toxic release, an explosion, a fire with off-site consequences. Generic continuous-improvement advice treats all findings as roughly equivalent inputs to a backlog. Prestatiebewaking requires you to weigh findings against major-accident scenarios specifically, which means your monitoring has to be structured around your risk assessment, not around a generic issue tracker.

The obligation is prescriptive, not aspirational. "Encourage front-line employees to suggest improvements" is good practice, but it isn't a compliance answer. Annex III asks for procedures - defined, documented, repeatable - for assessing compliance with MAPP objectives. A culture of continuous improvement doesn't substitute for a monitoring procedure that specifies what gets measured, how often, against what threshold, and who reviews it.

Informal tracking doesn't survive an inspection. PDCA run through spreadsheets, email threads, and paper checklists might genuinely produce better safety outcomes over time. What it won't produce is an audit trail. When a competent authority asks how you monitor performance against your MAPP, "we have a culture of continuous improvement" isn't an answer they can verify. They need to see the indicators you track, the frequency, the findings, and what happened after each finding - with dates and ownership attached.

Generic advice stops at "review regularly." It rarely addresses the harder part: distinguishing signal from noise across hundreds or thousands of inspection and audit data points, spotting a trend before it becomes an incident, and proving the loop actually closed. That's a data and process problem, not a mindset problem.

How This Connects to Inspection and Audit Software

Performance monitoring under Seveso III is, at its core, a data problem: you need consistent, timestamped, attributable records of inspections, audits, near-misses, and corrective actions - captured in a form that lets you see trends, not just individual entries.

This is the specific gap Capptions' Seveso Control is built to close. Digitizing inspection and audit workflows gives a site the raw material prestatiebewaking requires:

Structured, repeatable data capture. Digital inspection and audit checklists - tied to the safety-critical elements identified in your risk assessment - produce consistent records instead of the variable, paper-based documentation that makes trend analysis unreliable.
A live corrective-action loop. Findings from inspections and audits are logged, assigned, and tracked to closure, so "feeding results back into the management system" is a built-in workflow step rather than something that depends on someone remembering to circulate a report.
Trend visibility across sites and time. With inspection and audit data centralized, patterns that would be invisible in a filing cabinet or a shared drive - a specific asset type generating repeat findings, a site consistently missing inspection deadlines - become visible enough to act on before they turn into an incident.
An audit-ready record. Every inspection, finding, and corrective action is dated, attributed, and retrievable, which is exactly the form of evidence a competent authority expects to see when they ask how you monitor VBS performance during a Seveso inspection.

None of this replaces the analysis and judgment your EHS team applies to the data - deciding which trends matter, which indicators to prioritize, how findings map back to MAPP objectives. What it does is remove the administrative gap between "we're supposed to monitor performance" and "we can show exactly how we do."

The Bottom Line

Prestatiebewaking isn't a mindset - it's a documented, evidence-producing process for checking whether your safety controls are actually working, and for acting when they aren't. Generic continuous-improvement thinking can inform how a site approaches that process, but it doesn't satisfy the Annex III requirement on its own. What satisfies it is a monitoring system built around your specific MAPP objectives and risk-critical elements, running on data that's structured, trended, and traceable - which is exactly what a digital inspection and audit platform like Seveso Control is designed to support.