What is Learning from Normal Work?
A modular system that helps organizations find the causes of future incidents in the work that is happening right now, before something goes wrong.
Five problems the industry could not solve with existing approaches
-
Incident investigations ran out of incidents to learn from
Injury rates dropped, which meant fewer investigations and less visibility into how risk was developing. Repeat incidents kept occurring. Learning only from failure was no longer sufficient.
-
Risk assessments missed half the picture
Most risk assessments focus on hazards and energy sources. They do not capture error traps: confusing controls, ambiguous procedures, tools that do not fit the task.
-
Behavioral observation programs plateaued
Organizations running BBS for years reached a point where most observations were "safe" and the program stopped adding value. Correcting behavior without addressing the conditions behind it produced diminishing returns.
-
HOP philosophy did not lead to action
HOP gave the industry powerful ideas. Safety II added the insight that failure and success have the same causes. But organizations that trained leaders in these principles found that understanding the philosophy did not produce improvement or integration with the business. Both remained conceptual.
-
Blame was destroying the information flow
Leaders responded to failure with punishment. Workers responded by hiding what really happened. The organization lost access to the information it needed most.
From idea to practical methodology
If failure and success have the same causes, the conditions behind your next incident already exist in the work that is going well today. You can find them before anyone gets hurt.
Learning from Normal Work evolved from an initial set of ideas (IOGP Report 642) into a modular system of structured methods and tools through systematic application across industries.
You cannot use root cause analysis on work that has not failed. If you ask a worker doing their normal job "why did you do this?" the answer is "I am just doing my job." Questions designed for failure do not work when there is no failure.
Why the methods had to change
Five minutes on how it works
Dr Marcin Nazaruk introduces Learning from Normal Work: what it looks like on a site, what it asks of leaders, and what it changes.
Four layers of change
Mindset
What your people notice and pay attention to
People with the same information, but different mindsets, reach different conclusions. Without changing the mental model, introducing tools and processes becomes ineffective and shortlived.
- Safety is the lack of accidents
- Safety is the capacity to succeed under varying conditions.
- Unsafe behavior is the main cause of accidents
- Unsafe behavior is a symptom of trouble deeper in the system.
- Leaders should catch and punish
- Leaders should listen and address what makes work difficult.
Skillset
What your people can do about it
Learning how people actually work requires them to admit errors and workarounds, which takes trust and exploratory questions rather than verification questions that only confirm the checklist.
- "Why did you do this?"
- "What is the advantage of doing it this way?"
- "People disappear when I visit."
- "People open up about their challenges."
- "We punish first to teach a lesson."
- "We need to learn first. Trust is essential."
Toolset
What structured methods they use in the field
Our LFNW tools range from 10-minute learning conversations that fit inside a leadership visit to multi-day pre-accident investigations, making application repeatable and measurable at scale.
- Behavioral observations correcting behavior
- Learning conversations understanding constraints.
- Leadership visits
- Learning conversations discovering constraints and challenges.
- Risk assessments focused on hazards
- Dialoges about hazards, constraints, adaptations, and dependencies.
Scale
How they integrate it into existing processes
Integration means incorporating proactive learning into the processes your organization already runs:
- Leadership visits
- Risk assessment
- Investigations
- Behavioral observations
- Accountability
- KPIs
The system becomes part of your existing management system.
How it relates to HOP and Safety II
HOP gives you the principles: people make mistakes, blame is counterproductive, and context shapes behavior.
Learning from Normal Work turns those principles into structured tools your teams apply in the field, with the training that builds the mindset and skills to use them and the process integration that makes the change stick.
Safety II gives you the perspective: study how work succeeds, because success and failure share the same causes.
Learning from Normal Work supplies the methods to study normal work in practice, so the perspective produces findings your organization can act on.
Evidence from the field
58% 91
Leadership questioning quality
Measured before and after training (p<.001).
TE Connectivity
37
Injury reduction over 18 months
Published in Professional Safety (ASSP).
Professional Safety, ASSP
50
More hazards identified
After a single training session.
$500K
Cost avoided from a single finding
A Competency Pathway participant caught a specification error before it spread.
Kenvue
Want to explore what this could look like in your organization?
If you would like to explore what Learning from Normal Work could look like in your organization, we are happy to talk through your situation.