Do you have 'confidence' in the investigation process?

Throughout my career I have been fortunate enough to be involved numerous investigations, ranging from criminal prosecutions to offshore oil rig fatalities. I have also had the honor to lead several of those investigation teams, often during very trying times for all involved. The integrity of some of my team members over the years has been nothing short of humbling.

As an investigation leader, you quickly get used to the fact you won't always be able to pick all the members of your team. This is because interested parties often (and understandably) insist that they have their own representatives within the group. Due to this, I always cover certain topics within my opening meeting with the investigation team, some of which I am meeting for the first time. This works very well so that we can all agree the 'ground rules'.

There are quite a few topics on the menu of that conversation, such as the philosophies that we will use when investigating. This includes addressing the moral and ethical standards that we will (without question) uphold whilst conducting our business. This is essential for many reasons, not least due to the potential of following the evidence into areas which can expose those in positions of authority (such as our employers) to criticism. On the rare occasion that this happens, even the most robust moral compass is in danger of being knocked off course.

This conversation also gives me the chance to gauge competence, ensure the protection of critical processes (chain of evidence, etc.), discover any predispositions a person may hold and agree the need for transparency in our confidence levels regarding our interpretation of a) the evidence and b) what we believe it is telling us. The latter point is the subject of this article, and something I believe is rarely addressed during incident investigation courses.

For context, Scots Law requires the burden of proof in criminal cases to demonstrate guilt 'beyond reasonable doubt' (high confidence); whereas civil cases only need a 'balance of probabilities' (51% decision). Now imagine the different outcomes from an investigation process based only on absolutes (as is needed in criminal law); compared to that which also includes the 'most likely' (seen in civil cases). Could we say that the two approaches would get the same root causes, or would the inclusion of the 'probable' dilute the process just too much? There is certainly no shortage of case history where this very thing has happened.

Therefore, when team members input their conjecture into the process, it is important that they also feel comfortable enough to give the degree of confidence that they have in their offering. The group then uses that information within their decision criteria on the direction that they take the investigation. This is important as an investigation may take a completely different turn when based on a definitive conclusion, compared to a working hypothesis.

However, the leader must also appreciate that the reader of the final report must also be given an indication of the confidence levels of the investigation team, and the rationale behind the decisions that were made. This is so that the report's final recommendations can be validated, which is a critical (and often missing) step in getting them implemented, and especially relevant when they involve a ($) resource allocation or organisational change.

In summary, an incident investigation report is simply one of many business improvement tools that a leadership team will use to base their strategic decisions. For that reason, don't be discouraged if a report that contains no more than a fishbone and some recommendations is rejected and handed back to you with a request for more context. Just like your teacher said at school, it pay dividends to show your working (that's where you get the best marks).