Analysing Incidents

Analysing Incidents

The art of analyzing incidents to understand exactly what the root cause and dynamics were, what could and couldn’t have been done, what was right and was an error, and what outcomes were preventable and what not, is not easy.

To do this requires, first, a good understanding of the people and the activities and circumstances around the incident.  It means really knowing a client well.

Then it requires a thorough working knowledge of the incident lifecycle, a topic of much research of late.  You need to understand how incidents play out and the patterns you tend to see repeating in incident occurrence and response.

Then to this you have to add a solid ability to apply definitions and categorization systems.  You have to be able to know what is being defined as an incident, a near miss, a preventable adverse effect and many other incident specific terms.  You need to know this in the context of the industry, geographic location and specific activities of the entities involved in the incident.

After that you need to bring an encompassing awareness of human behavior, cognitive psychology (which is the field of psychology interested in how we think, process information, make decisions an

And lastly you need to know statistics, data and have experience in real world incidents, the more the better.  Without this solid empirical background it is easy to get misled by impressions and biases within yourself.  You need to know what really is linked to what in a  causative fashion and what really underlies well known behavior patterns. And often what we think causes something is wrong and someone somewhere has proved it. If you don’t understand how people can make mistakes and have a working taxonomy of error which defines the classes of mistakes possible then you won’t be able to see them.

Incident analysis can be very powerful as a learning tool.  It can genuinely help guide preparedness and training and can help you move from a state of powerlessness to a state of empowerment.  It does matter.  But it isn’t easy and if you are given the wrong “insights” then your decisions and often your expenditure won’t be effective, you will be placing your ladder against the wrong wall.

Because of this we don’t usually recommend clients try analyze their own data.  There is too much that can go wrong with the inferences that will be drawn.  We strongly advise them to contact us and seek external help on this.

Lastly bear in mind that not all incidents lend themselves to analysis and not all analyses are useful.  Taleb, in his seminal work on the unpredictable, covers this quite nicely.  As he put it; what we learnt from the 9/11 disaster in the USA are specific lessons to prevent fundamentalist Islamic terrorists flying jet airliners into tall buildings.  If we think we learnt generally useful lessons from the enormous amount of analysis work that has gone into that incident, we are wrong.  It was an outlier, an unusual event.  And unless we start seeing it replicated over and again in future terrorism incidents it is a bit of a statistical freak event.  The lessons therefore aren’t going to help us.

So don’t get too locked onto analyzing everything, we can guide you as to what can and can’t be achieved and can help you get really powerful and relevant insights that do make a difference.