Why Critical Control Verifications Turn Into Tick-Boxes

Practical Steps You Can Take For Trustworthy Verifications

Key takeaway: When critical control verifications get rushed or completed in bulk, it’s rarely because people don’t care. It’s often a sign the verifications are too long, too frequent, too easy to answer without doing the test, or that nobody looks at the results. Fix the design and the feedback loop, and your verifications start telling you the truth again.

It’s 4pm on a Friday. A site supervisor opens the inspection app and works through a month of overdue verifications – emergency stops, machine guarding, exclusion zones – in under four minutes. Every verification comes back as a pass, and on the dashboard the site turns green.

Anyone who’s run an inspection or pre-start programme at scale has seen some version of this – say, a situation where verifications designed to take 15 minutes are regularly completed in one – what some in the industry call pencil whipping.

It’s tempting to treat this as a discipline problem. More often, it’s a design problem – and that’s good news, because design is something you can change.

Why a tick-box verification is worse than no verification

A missed verification is visible – it shows up as overdue, and someone can follow it up. A verification that’s ticked but not done is invisible. It tells leaders a critical control is in place and working when nobody has looked, and for critical risks, where a critical control failing can mean someone doesn’t come home, that’s the most dangerous kind of false comfort.

A completed checklist is a record that someone said the work happened. Whether it happened is a separate question, and it’s the one officers are expected to be able to answer – so you need to know you can trust verifications have been done genuinely.

The IoD’s Health and Safety Governance guide puts it neatly. Boards should “fear the green and embrace the red” – and be curiously sceptical when every indicator comes back positive. A wall of passes isn’t proof that critical controls are working. Sometimes it’s proof that nobody is testing them properly.

Why verifications get rushed

When you look closely at verifications that have become tick boxes, the same causes come up again and again:

• The verification is too long. A 30-question form for a five-minute task invites skimming, and the questions about critical controls get buried among the trivial ones.
• The questions can be answered without doing the test. “Is the emergency stop functional?” can be ticked from the smoko room. Nothing about the question requires anyone to press the button.
• The frequency doesn’t match the work. Verifications set by the calendar rather than by how often a control is relied on end up either too thin or so frequent they become background noise.
• Nobody looks at the results. If a failed verification disappears into a system and nothing happens, people quickly learn the verification is paperwork.
• Bad news isn’t welcome. If reporting a failed control leads to blame or a pile of extra work, a pass becomes the path of least resistance.
• There isn’t time. Verifications squeezed into the last 10 minutes of a shift will be done the fastest way possible.

None of these is about people not caring. Each is a signal that the system is asking for something people can’t reasonably give, or isn’t giving anything back.

What helps

The fixes follow from the causes. None of them needs a big programme – most are changes to how verifications are written, scheduled and followed up.

• Write each verification as a physical test. Instead of “Is the emergency stop working?”, ask “Activate the emergency stop. Did the machine stop?” A question that can only be answered by doing the test is much harder to tick from a distance.
• Keep critical control verifications short and separate. Pull them out of general housekeeping forms so the questions that matter most aren’t buried.
• Tie frequency to how often the critical control is relied on, not the calendar. A daily verification on a machine used twice a week becomes noise, while a weekly verification on equipment used hundreds of times is too thin. A practical starting point is the 10% rule of thumb – verify once for every 10 times the critical control is relied on. Scale back only once results are consistently positive, and only as far as the control’s stability and the pace of change in the work allow. Less stable controls need more frequent verification, whatever their recent track record. We cover this in more detail in how to set critical control verification frequency.
• Ask for evidence in proportion to the stakes. A photo of the isolation lock in place, or a short video of an interlock tripping, makes the verification real for the person doing it. Keep it for the critical controls that justify it, or evidence becomes another box to tick.
• Look at patterns, not just results. Overdue verifications completed in one sitting, inspections finished impossibly fast, or a site that has never recorded a single failure – each is a question worth asking, not an accusation.
• Review the verifications. IoD/WorkSafe governance guidance describes three layers of good verification practice: regular proof testing at point of use by those who manage the risk, periodic higher-level reviews to make sure that testing is working, and system audits of the whole approach. One of the simplest ways to do the second is for a supervisor to occasionally do a verification alongside the person who normally does it. It helps if that second layer sits outside the line being verified. When the person verifying reports to the manager whose site it is, a red result can feel like a career risk – nobody should be marking their own homework.
• Respond well to a fail. Thank the person who found it, fix the critical control and close the loop so they can see what happened. Every well-handled fail makes the next honest one more likely.

Aim for visibility, not surveillance

This is the line to hold. The goal is to make it easy to see whether verifications are genuine – not to time every worker to the second or rank people by how fast they go. Verifications designed to catch people out tend to produce people who are better at not getting caught.

What changes behaviour is visibility paired with the right conversation. When patterns are visible to supervisors and leaders, “help me understand how this month’s verifications all got done in four minutes” lands very differently from a disciplinary letter. The point isn’t to tell anyone off – it’s that everyone has their part to play so we all get home safe.

The question to be ready for

The question your board should be asking – and you should be able to answer – isn’t “are our verifications being completed?” It’s “what is our confidence in these verifications based on?”

If the honest answer is a completion rate, there’s work to do. If it’s well-designed tests, evidence where it counts, a second layer that reviews the verifications and a team that’s comfortable reporting a fail, you’re in a strong position. Your green means something.

Part of our critical risk governance series.

Critter is built by IMPAC – New Zealand’s leading health and safety company, with 27 years of experience guiding organisations through complex critical risk challenges. Learn more about IMPAC.