Assessing risks: a different method (4/4)

In my series on risk assessment, I explained how most organizations approach their assessment. In the fourth post in this series, I propose another method that differs from the well-established risk management method: scoring the probability and impact of a risk on a (quantified) scale of 4 or 5. Not everyone handles this well and it is difficult for many people to estimate the absolute probability or impact. Scrum has realized for a long time that this is not optimal and are fans of relative estimating. Scrum.nl explains why:

As an example, we can hardly estimate how much a car weighs and how much a scooter weighs in kilos. Or another example: if we see two towers, we can see at a glance that one tower is twice as high as the other. But we cannot estimate exactly how high each tower is in meters. In other words: we can hardly estimate in absolute terms, but we can estimate relatively quite well.

Putting it into practice

I was wondering why could this not work for risks as well? How this works:

1. You make an inventory of all your risks (as you normally would)

2. Then you do not estimate them on an absolute scale, but relative to each other. So a risk with a higher probability will be higher in the list, or will receive a higher grade, whichever order you prefer. You do not have to argue whether it is 3 or 4, you just have to consider whether it is greater or smaller than another risk. If you estimate probability and impact, you end up with two lists.

3. You add up the scores to arrive at a risk score (multiplication is also possible). The risks with the highest / lowest score are the ones that you should pay the most attention to. An example (I chose the higher the probability/ impact, the higher the score):

The risks of fire and decay have the highest final score and are therefore the risks that I have to care the most about. Another person may score differently on e.g. impact: this also changes the final score. It remains a personal assessment: I therefore always recommend doing this with several people, especially in larger organizations. In this way you create a supported score that accurately reflects the average estimate.

Have you taken sufficient measures against fire (such as fire-resistant materials and smoke detectors) and decay (for example, a maintenance contract with a contractor)? If you desire so, assess the risks again, scoring the probability and impact and take into account the measures taken. Are they dangling at the bottom of your list? Then the estimate is that they work well. Logically, 2 or 3 other risks suddenly will become high: that score is not all of a sudden different, but relatively speaking it has risen compared to the controlled risks. So do not be alarmed by this.

A final tip for this application: make sure that your top list of risks does not become too large: for example, if you have identified 14 risks, tackle 3, max 5 at the same time. With this you can apply focus and really control that small top well, instead of a top 10 that you only half control. That is a waste of your time and money.

More possibilities

This method can also be applied in a different way. In the mortgage market (both for private homes and commercial properties) a lot of customization is done for certain types of loans. Often a loan application is prepared by an advisor (internal or external) and then assessed by an internal credit analyst (sometimes called a risk analyst). In my experience, the approach to assessment can differ enormously between analysts.

Often there is no standard method to perform an assessment because it concerns customization to which no standard procedure can be applied. Where one analyst only looks at sufficient collateral (such as collateral, or guarantees, regardless of the other content of the application), the other analyst considers all aspects and only approves when all aspects are at least sufficient. The advantage of this is that each application receives a tailor-made analysis, the disadvantage is that the outcome is highly dependent on which analyst makes the assessment.

Relative estimation can provide a middle ground between complete freedom and a standard procedure. This can be done in a number of steps:

1. Determine a number of aspects in advance that are relevant in each application and that could pose a risk. Example: income of the applicant (type, consistency, height in relation to costs / LTI, etc.), location of collateral, type of collateral, collateral application options, collateral maintenance statement, other securities (such as a surety or pledge of rental income), amount of the requested loan compared to the estimated value of the collateral (LTV in professional jargon), term fixed-interest period, etc.

2. Draw up a standard form on which each assessment must be completed, leaving room for a customized assessment.

3. Start each assessment with a relative risk assessment of all aspects of the specific loan application. Herein the aspects listed under #1 are scored relative to each other with the highest risk coming first. And then set the rule that, for example, with a list of 15 aspects, the top 5 must be assessed extensively, since they pose the greatest risk in this specific application.

The advantage of this method is that there is a standard method on which each analyst performs assessments, while at the same time there is a lot of room to maneuver to deliver a tailor-made analysis that is tailored to the individual request.

I'm curious what you think about it. Do you see opportunities for relative estimating in your work? Let me know!

Images through Unsplash