Stop Ticking Boxes: Turning Minor Incidents Into Real Organisational Learning

An underground vehicle rolls back and hits a concrete pump. The investigation, a full ICAM, concludes the worker failed to follow the procedure. Written warning, retrain, close the file. Twelve months later, at the same time of day, same task, a different driver, the same thing happens again. The first investigation changed nothing.

In Session 2 of the myosh Academy Investigation Series, Mark Alston of Investigations Differently used that story to make a simple point: the way most organisations handle minor incidents guarantees they learn nothing. This article distils his alternative,  Event Insights, and the seven questions behind it.

Why 'failure to follow procedure' teaches you nothing

Alston borrows a line from Todd Conklin: blaming an incident on a worker who failed to follow a procedure is like blaming a falling apple on gravity. It's always true, and it's useless. Procedures only ever describe the good day, they never include a step where someone gets hurt, so 'failure to follow procedure' is the most obvious and least helpful finding available.

“Our procedures always have happy endings. So the worker failing to follow one is a symptom, not a cause.”

The popular tools make it worse. The 5 Whys, designed by Toyota for production-line faults, assumes a single, linear root cause and, poorly applied, usually ends in blame. Predetermined tick-box lists ('worker not trained', 'poor situational awareness') simply confirm what you already assumed. You never learn anything new.

A better objective: learn and reduce risk

Alston argues the whole objective is wrong. 'Prevent recurrence' is a unicorn, the only way to guarantee it is to eliminate the activity. Delete every admin, training and PPE action from your last two years of investigations, he suggests, and see how little is left. The real job is to reduce risk, and risk is simply the effect of uncertainty, the things you don't yet know.

“My goal is to discover the things I don't know that I don't know. Because that's what risk is.”

The seven questions of Event Insights

Event Insights is built for minor and medium-harm events where there's a genuine opportunity to learn. It uses seven 'generative' questions, conversation starters, not an interrogation, designed to shift thinking from who failed to what failed. Three are about the event: describe what happened; what surprised you; could it have gone worse, and why didn't it? Four are about the work: when this task goes well, what must go right; what frustrates you; what could management better understand; and how could we improve how we do this?

The magic is in the framing. Asking what a good day looks like surfaces what workers actually need to succeed. Asking about frustration reveals where people deviate. And asking what management doesn't understand uncovers risk that never reaches the corner office. Best of all, it can be run collaboratively, as a group, out where the work happens, within a couple of hours of the event.

What this means in practice

Once you've gathered the themes, Alston offers a simple three-step process for minor events: apply the hierarchy of control (start at elimination, not admin), focus on solutions within the team's control, then choose what's practical and effective. The test throughout is whether the action genuinely reduces risk — an unguarded machine needs a guard, not another line in a 50,000-line procedure.

For organisations using myosh, this is where the platform earns its keep, capturing conversational, narrative data instead of tick-boxes, and (with an approved, sandboxed AI tool) pulling the recurring themes out of many Event Insights conversations at once.

Key takeaways

• 'Failure to follow procedure' is a symptom, not a cause; procedures only describe the good day.

• 5 Whys and tick-box lists confirm what you assumed; they rarely teach you anything new.

• The objective isn't to prevent recurrence; it's to learn and reduce risk.

• Event Insights uses seven generative questions to shift from 'who failed' to 'what failed'.

• Run it as a conversation, collaboratively, out where the work is done.

• Only use it where there's a real opportunity for organisational learning.

Watch the full session

There's far more in the full hour, including the underground mine story, the shark-attack analogy for hindsight bias, and Mark's three-step approach to controls. Watch on demand.  

This session is part of the myosh Academy free webinar series. Explore the library and register for the rest of the Investigation Series.

Presenter: Mark Alston, Founder, Investigations Differently. Tags/Categories: Workplace Safety, HSEQ, WHS, Incident Investigation, Learning Teams. Internal links: 'myosh Academy webinars' → https://www.myosh.com/academy/webinars ; related — Session 1 'Stop Investigating Everything'. Suggested image alt text: 'Supervisor having a learning conversation with a worker at the job site'.

‍