Near Misses: Getting the Reports That Actually Get Filed

Every site manager has the same story: the near-miss book has three entries this year, and two of them were filled in by the same person.

It is not that nothing is happening. On a busy manufacturing site, something is always nearly happening. The silence is not a sign of safety. It is a sign that people do not think it is worth the bother.

What the Ratio Tells You

Frank Bird’s triangle from 1969 is still worth understanding. He looked at 1.75 million incidents reported by 297 companies and found that for every serious injury, there were roughly ten minor injuries, thirty property damage incidents, and six hundred near misses.

The numbers are not gospel. Every site is different. But the shape of the thing is right: serious events sit at the top of a much larger pyramid. If you are getting one near-miss report a month, you are not running an especially safe site. You are running a site where people have stopped filing.

The ratio is useful because it gives you a target. If you have a hundred people on site and you are not seeing at least a handful of near-miss reports per week, something is wrong with the reporting, not with the hazards.

Why People Go Quiet

Ask anyone who has worked on a shop floor and they will give you the same list.

They think nothing will come of it. Someone filed a report two years ago about a slippery patch near the press. The floor got a cone on it for a week, then back to normal. Word gets round.

They do not want to look incompetent. Reporting a near miss means admitting you were in a situation that nearly went wrong. That takes a certain amount of trust in how the information will be used.

They are not sure what counts. “Nearly dropped a box” does not feel like a reportable event. Nobody told them otherwise, so they say nothing.

The third one is the easiest to fix. A clear, simple description of what a near miss is, posted where people can see it, removes a lot of the ambiguity. The first two take longer, because they are about culture.

Making It Safe to Report

The only thing that genuinely shifts reporting rates over time is what happens after someone submits a report. If the answer is nothing, or a form-filling exercise, or a quiet word from a supervisor about wasting time, reporting will dry up. If the answer is a visible change, or at least a visible conversation about why a change was or was not made, it builds.

A few things that work in practice.

Close the loop quickly. If someone reports a hazard, acknowledge it within 24 hours. Even if the answer is “we looked at it and here is why we are not changing the process right now,” people feel heard.

Do not turn reports into investigations of the reporter. Near-miss reporting should be about the system, not the individual. The moment it becomes about what someone did wrong, you have lost the next hundred reports.

Make it easy. Paper forms on a clipboard in the break room have a poor hit rate. A simple digital form that takes two minutes is better. Some sites use QR codes posted near common hazard areas.

Review near misses openly, not behind closed doors. A short slot in the monthly team meeting, going through what was reported and what happened next, does more for culture than any poster campaign.

What to Do With the Data

The reports are only useful if you look at them as a pattern. One near miss near the palletiser might be a one-off. Four near misses near the palletiser over three months is a process problem.

Keep a simple log. Date, location, type of hazard, action taken. After a few months you will start to see clusters. Those clusters are where your serious incidents are incubating.

This connects directly to how you think about losses more broadly. A near miss that does not become an incident still cost you something: the close call, the stress, the interrupted workflow. Mapping that alongside your other operational losses gives a fuller picture of where effort is being wasted. If you want a structured way to think through where the losses are on your site, the tool at losses.cadencestandard.com is worth half an hour of your time.

The physical environment matters too. A cluttered, poorly marked site produces more near misses because it creates more opportunities for things to go wrong. A site that has gone through a proper 5S exercise tends to surface hazards during the sort and set phases, before they become incidents rather than after. The guide at 5s.cadencestandard.com covers that in detail if you are starting from scratch or trying to revive a stalled programme.

Near-miss reporting is not complicated in principle. It is just one of those things that requires consistent attention to do properly, and falls apart quickly when it does not get it.