Zero injuries, zero information

It is telling them less than they think, and possibly the opposite of what they think.

• A zero count is a lagging outcome measure. It records what happened to people, not what is being controlled. With a workforce of 140 and three accidents a year previously, zero is statistically fragile - it can be produced by luck and by under-reporting as easily as by good management.
• It is contaminated by the reporting culture. Near-misses falling from 42 to 9 in the same period that a supervisor calls reports 'paperwork' and a fitter is laughed at is not improvement. It is the warning system being switched off. Fewer reports after public ridicule is the most predictable signal in the whole pack.
• It contradicts the evidence on the floor. Fourteen of twenty lifting items overdue for examination and three unguarded press brake returns mean serious-injury potential is present and uncontrolled while the headline says 'under control'. Zero harm so far is luck, and luck is not a control measure.
• The data are migrating. LTI frequency falling for three years while MSD-related absence rises 20% suggests injuries have not gone away, they have moved into the sickness absence figures where nobody is counting them.

The board is being reassured by a number that measures silence. The right response is to challenge it, not print it.

Keep the headline for continuity, but stop letting it stand alone. A workable set for this size of business:

Reactive (what happened): RIDDOR-reportable injuries and lost-time injuries as now; plus lost-time absence split by cause, so MSD cases cannot hide in the sickness figures.

Proactive (what is being controlled):

• Percentage of statutory thorough examinations completed by the due date (currently 6 of 20: a single number that would have stopped the board in its tracks).
• Critical control verification: guarding confirmed present on the press brakes at each pre-use check, reported as a percentage of checks done.
• Corrective action closure rate and average days overdue - the 60% figure made visible and trended.
• Near-miss reports per 100 employees, and the proportion closed with feedback to the reporter.
• Induction and refresher training completion for safety-critical roles.

The balance is deliberate: at least one measure of harm, several measures of exposure and control, and at least one measure of reporting health. Proactive measures get targets; reactive measures get trends and investigation, never targets, because a target on injuries rewards silence. If the leading and lagging indicators disagree, believe the disagreement and go and look.

Open with the shared goal: twelve months without a reportable injury is genuinely worth having, and the concern is that it is fragile rather than fake - the organisation is one unexamined lifting sling away from a very different board pack.

Then triangulate with three simple charts and one sentence each: near-miss reporting falling while a supervisor dismisses reports; thorough examinations overdue and climbing; corrective actions slipping past their dates. Add the absence data so the MSD rise is visible. Keep it to a page - the argument is the pattern, not the detail.

Translate into the board's own language: an overdue lifting examination or an unguarded brake is an uninsured event waiting to be priced, and the cost of a serious injury or an enforcement notice dwarfs the cost of clearing 14 examinations and three guards.

Three decisions to ask for:

1. Name an owner and a monthly one-page report of the leading measures to this board.
2. Resource the immediate clearance of the overdue examinations and the guarding, with dates.
3. A visible, repeated message from the top that reporting is wanted and will not be punished.

And handle the supervisor as part of the solution, not the exhibit - the reporting culture is a management product, and humiliating him will teach the same lesson he taught the fitter.