Root-Cause "What," Not "Who": 5 Investigation Habits That Keep Your Team Talking

Why "who did it" is the wrong opening question

Blame feels efficient because it closes the file fast, yet it answers nothing about why the action made sense to the person at the time. On Headline Podcast, Tim Page-Bodoff put it plainly: "I don't think we should do a root-cause who. I think we should do a root-cause what." The distinction matters because a named culprit ends the inquiry exactly where the useful part begins. OSHA reports that struck-by, caught-in, fall, and electrocution events, the construction "Focus Four," account for more than half of construction fatalities, and almost none of those traces back to a single careless individual rather than to layered decisions about layout, scheduling, and supervision made long before the shift started.

The accident is a construction, not bad luck

The deepest reframe comes from the host's own work. As Andreza explores in her book Sorte ou Capacidade (Luck or Capability), an accident is systemic, the late result of layers and barriers that failed in sequence, what James Reason described as the Swiss cheese model, not a random stroke of misfortune. Her companion line from Um Dia Para Não Esquecer sharpens it: "It wasn't chance. It was construction." When investigators internalize that an event was built, decision by decision, over weeks or years, they stop hunting for the unlucky operator and start mapping the holes that lined up. ISO 45001 specifies, in its incident-investigation requirements, that organizations determine the underlying causes and act on the management system, not merely the immediate trigger.

It always looks like the employee's fault until you look deeper

There is a predictable pattern in serious-event analysis that explains why blame is so seductive. On Headline Podcast, Dr. Thomas Krause noted that decision analysis of serious events "always looks like it was the employee's fault" for not following the procedure, until you look deeper and find that following it was made nearly impossible by system factors set by decisions made a month, a year, or five years earlier. The procedure existed. The training record was signed. The PPE was issued. And the work was still designed so that the safe path was the slow, awkward, career-limiting one. NIOSH recommends prevention-through-design precisely because the cheapest, most durable barriers are built upstream, long before the worker reaches the task.

Turn the finding into something the line actually owns

A root-cause "what" only pays off if the fix lives where the work lives. The practitioner rarely owns "what do we do about this," so a corrective action handed down from the safety office and never adopted by the department that runs the operation will not last even when everyone nods in agreement. The most reliable investigations end with the line manager and crew co-authoring the change, because the people who run the work hold the knowledge no procedure captures. BLS records tens of thousands of recordable cases each year where the recurrence, not the first event, signals that the original corrective action never reached the floor.

Don't wait for the accident to be your teacher

The worst place to learn about a risk is from the accident itself. Mature operations discover risk proactively through structured inquiry, treating a near miss as an accident that has not happened yet rather than a non-event to file and forget. As Andreza frames it on the show, a workplace accident is a book we did not read, and the near-miss log is usually the table of contents nobody opened in time.