You fell and hurt your arm. X-rays reveal that it’s broken, and you’re in a lot of pain. What would you think if the doctor gave you pain medication and sent you home, without doing anything else? Speaking personally, I wouldn’t have a very high opinion of the doctor. The pain is just a symptom and treating it is a temporary fix. The broken bone is the root cause. If it isn’t addressed, the symptom will persist and the problem itself will get worse over time. Training solutions can fall into this same pitfall. If they don’t address the root cause of a performance issue, they won’t resolve it or have lasting positive impact.
I’ve given an example of root cause analysis before, so this time let’s dive deeper into the “how to”.
Note that with any root cause analysis you’ll need access to data, feedback, and/or other information. You might be working with existing information or need to shift gears part way through the analysis to gather what you need. Either way, the outcome’s quality is partially dependent on how accurate and complete the information is. Ideas and hunches can be useful, but they don’t generally have the same weight as working with hard facts.
Here are some common methods for conducting root cause analysis.
Root Cause Analysis Method: The 5 Whys
This is probably the best-known method, popularized by the case study of a monument in Washington D.C. eroding. The theory is simple: you keep asking “Why?” until you arrive at the answer (the “5” shouldn’t be taken literally).
One of the shortcomings of this method is that it provides little guidance. “Why?” is a broad question and it can be difficult to take all of a situation’s variables into account. If it’s too open-ended for you, consider using one of the other methods. They still involve digging to get to the root cause of a problem but provide more structure.
Root Cause Analysis Method: Root Cause Tree
A root cause tree is a visual representation of the causes and effects involved in a problem. When it’s complete, it looks like the roots of a tree. Start by putting the problem at the top. Then, use arrows to work your way down through the symptoms, possible root causes, and likely/actual root causes. Each symptom should be the start of a new “tree root” that follows its way through the possible and likely/actual causes for that symptom. Organizing the variables this way makes it easier to follow the sequence of events and identify relationships.
Root Cause Analysis Method: Fishbone Diagram
A fishbone diagram is another type of visual representation. The basic concept and setup are like a root cause tree, except for two main differences. First, the diagram is oriented from side to side, instead of from top to bottom. When it’s complete, it looks more like a fish skeleton than tree roots. Second, it’s organized by factors rather than symptoms. The factors involved will vary depending on the type of problem. Examples include people, processes, equipment, location, training, and so on. Possible root causes from each factor are then added as “smaller bones” coming off the main factor lines. Organizing the variables this way makes it easier to see which factors are causing the most difficulties.
Root Cause Analysis Tips
No matter what method you use to map out possible root causes, you still need to figure out which of them are “the real deal”. Sometimes a problem has a single main cause. Other times, it may have several significant contributing causes. Here are some tips for making decisions during the analysis and resulting decision making:
- Would addressing a specific cause resolve more than one symptom?
- If so, it’s probably a good idea to prioritize that solution.
- Is complete resolution a realistic goal or is improvement more likely?
- Sometimes a complete resolution isn’t possible or would be very expensive or time consuming. In those instances, you may have to pick the most practical solution over “the best” solution.
- What if addressing the root cause you identified doesn’t fix the problem?
- Either it wasn’t addressed very well, or it wasn’t a true root cause. You may have to investigate a bit more to figure out which is the case. If the chosen solution wasn’t applied adequately, you can try again. If it seems more likely that it wasn’t the real root cause, re-evaluate your analysis and see if you can identify the real culprit.
Addressing the symptoms of a problem without treating the root cause doesn’t resolve the issue. As you can see, there are multiple methods for conducting root cause analysis, but finding a solution is the goal. The only way to truly know you’ve found the root cause is to watch the problem disappear once you’ve addressed it. So, what are you waiting for? Get digging!
Ready to find out what Digitec can do for you?