"Human error" as a root cause explains nothing, it just names who was standing closest when the system failed. A HOP-based investigation asks a harder question: what conditions made that error the reasonable, even predictable, choice at that moment? That distinction changes what the corrective action actually fixes.
This isn't a generic root-cause template applied to your incident. It's built from having managed operations where the same pressures, schedule, cost, production targets, were present, and knowing which of those pressures typically drive the behavior an investigation needs to explain.
Investigations that stop at the worker's action instead of the conditions that made it the reasonable choice
Corrective actions that amount to “retrain the employee” and nothing else
Root cause findings that would look identical no matter what actually happened
Near-misses that never get investigated at all because nothing bad happened this time
“Every incident that occurs was allowed by the systems and culture that existed before it happened. The question is never who failed. It's what failed around them.”— Core HOP Principle, Dr. Todd Conklin
The goal of every investigation is a recommendation that changes the conditions, not one that just closes the file.
That's the same principle behind every investigation Black Swan runs: the goal isn't to close the file, it's to change the conditions that produced the event, so the same failure doesn't get a second chance to happen.
If the incident has already produced an OSHA citation, the investigation and the citation response need to tell the same story. See how OSHA citation response works →
Every investigation follows the same sequence, built to explain the event rather than just document it:
Reconstruct the operational conditions present at the time, not just the sequence of the event.
Identify what the system made easy, and what a different person in the same conditions would likely have done.
Recommendations that alter the conditions, not just the paperwork.
This is a complete investigation engagement, not a template report. Specifically, that includes:
Investigation launch within 24 to 48 hours of notification
Interviews with everyone involved, not just the person closest to the event
A written report identifying system-level contributing factors
Corrective action recommendations ranked by impact
Optional presentation of findings to leadership or the safety committee
No, it strengthens it. Most operators already have a reporting mechanism. This is the investigation methodology applied once something's been reported, especially for events serious enough that “human error” isn't a satisfying answer.
Near-misses are often the more valuable investigation, since the conditions were present without the consequence. Treating them as the free data they are is one of the fastest ways to reduce recordables.
Traditional RCA typically stops at the human action. HOP asks what the system made easy or difficult, treating the human action as information about the system rather than the conclusion of the investigation.
Typically within 24 to 48 hours, sooner for anything serious. Evidence and memory both degrade fast, so speed matters.
The methodology is documented and defensible, but for anything with likely litigation exposure, we recommend looping in your counsel early, not after the report is finished.
Yes, when appropriate. Front-line workers often speak differently about conditions when a supervisor isn't in the room, and that difference is often the most useful data in the investigation.
You do. The findings and recommendations are yours to use, share, or act on however fits your organization, whether or not further engagement follows.
Every engagement starts with a direct conversation, not a proposal. Tell us what's happening and we'll tell you what we'd actually do about it.