Risk Assessments - fun, fun, fun.
The first problem is everyone says "do a risk assessment ...." And what they really typically mean is more of an
FMEA. They want you to consider what could go wrong, and take steps to prevent it.
A true assessment results in a score, that you can use comparatively between options and select the one of least risk. Unfortunately, there is no such thing as a canned, generic risk assessment.
I could do a risk assessment on Trump's cabinet picks OR I could do an assessment on boat shipments getting captured by pirates OR I could do one on will my new manufacturing line make the needed timing. And in all of these, the QUESTIONS I would ask would be totally different. So you can't just say "risk assessment" without any qualifies as to what it is actually assessing. "Trump's Cabinet Risk Assessment" has a qualifier and makes more sense.
So how do you do it?
Pretty much, you brainstorm a bunch of questions that need to be answered and assign a pass/fail or Likert scale (strongly disagree to strongly agree) and add up the score. Pay close attention that they all align - how you WORD the question may drive whether or not the high number is the good number or the bad number. They all need to be logically oriented the same way.
Here's the weakness with this - it then leads to all questions counting the same. Which may be very misleading. In my "Trump's Cabinet" assessment, one question may be "Are they addicted to drugs?" and another may be "Are they older than 70?" Clearly, the one is more important than the other..... But if their score counts the same, you may not get good discrimination.
So the next level is you then RANK the scores. Give them weight. You can do this by sitting around flinging weights out there. I find that this is not productive and time consuming. There's lots of arguing over the ranks. The alternative to this is you can use something called "The Pairwise Comparison" method.
Here's a link to how it works ...
The goal of this is to appropriately weight your factors. Basically, it works like this:
1) You come up with your questions/factors and put them in a matrix.
2) You use this matrix to step through your factors, comparing only two at a time. Why? It's usually easier to agree on "which is the most important?" when you are only comparing two things instead of an entire list.
3) This exercise then accumulates wins and losses in the comparison, and THAT then becomes your ranking. In other words, if "Addicted to Drugs" comes out as more important than most things on the list, it will have the highest rank.
