Analysis. Analysis. Analysis. The essence of the R C A process.

LET’S START WITH OUR CALL TO ACTION: when you hear someone say they are doing RCAs or “Root Cause Analysis” – challenge their claim! The vast majority of the time, the abbreviation or phrase is thrown around without regard to the actual essence of RCAs. 

Note that – in the title – we left spaces between the letters. Why? We want to provocatively dissect the art / science of Root Cause Analysis, and an abbreviation will not do. 

Let’s start with the A. Analysis. The dictionary.com definition of analysis is that it is:

Why start with the A? Why not the R? It is, after all – the first letter in R C A. Here’s the reason: if there is no analysis that has been conducted – then the investigation should not be called an R C A. Often – we see documents that are incident summaries that are incorrectly categorized as RCAs. Also – statements like “logs were reviewed and no cause was found” or “system was rebooted to fix the issue” are not deeply analytic, let’s face it. Now – if the user has conducted a five whys analysis, or done hypothesis generation and testing, or some type of event re-simulation – then can we claim the A is valid! So you can ask these questions:

• What analysis has been conducted?
• Have logs been reviewed (in the case of an IT issue)?
• Have schematics been reviewed / studied? 
• Have multiple hypotheses been considered? 
• Have procedures been reviewed? 
• Have interviews been conducted?
• Has the “therefore” test been considered?
• Has physical evidence been examined (for instance, in the case of physical failure analysis)?

Thus in dissecting R C A we have reviewed the "A so far.

Let’s move onto the C. Causality.  Reason being – if Analysis has been conducted – and no cause has been found then of course the “Analyst” cannot claim that an R C A investigation has been undertaken. It may seem overly obvious – but if causality is not established through fact and logical connection of events and not by coincidence – perhaps the C in R C A can be claimed. Causality implies that an event is the producer of an effect. In absence of this connection – there is not causality, and the claim of having an RCA is negated.

Now that we have critiqued A and C let’s move to the R, the Root of it all. The Root of the issue is the item that, if addressed, then the problem would not recur. 

So – in an interesting analogy – R C A is really A C R– Analysis based on Causality to find the Root of an issue, which emulates the 5-whys process in that it considers the sequence of events in a reverse manner. From an English language perspective this makes sense – since the noun is key – again, Analysis is the key word in this whole endeavor. 

In summary – three things really matter in RCAs – Analysis, Analysis, and Analysis. Once could argue that causality is key – but in essence it’s a form of analysis. RCAs must be structured with discipline with logic. Analysis and causality! Let’s do an R C A – or put another way, let’s do A with critical C linkage to determine R!  Thinking of these items consciously and critically will make you a better problem solver!

This is the first in a series of articles on Root Cause Analysis - where we will look at RCA ecosystems, best practices, and case studies. Please send us a message if you want to join our LinkedIn RCA group. Thanks for your support!

-      Alan Piciacchio & Thomas Snellgrove