Why Red Amber Green

A Key Performance Indicator (KPI) is meaningless unless its value can be compared to something. Without a comparison it is impossible to say if the value is good, bad or indifferent.

The comparator could be a target based on previous performance or on a notional future performance or an industry benchmark or even a made-up value. Whatever the target, it needs to be reasonable and achievable.

Targets are well understood when looking at financial measures; we often look at a ‘variance’ to an expected result. For example if expected monthly revenue is £325k and the actual revenue recorded is £309k the variance would be -£16k. This may or may not be a cause for concern depending on what was considered an acceptable variation to the target.

For a KPI to be useful acceptable and unacceptable results need to be clearly understood. This are usually described as defining ‘thresholds’.

There are several threshold models available, for the purpose of illustration let’s look at the most common: Red Amber Green (RAG). In the RAG model there are two threshold points:
• When the KPI should turn Green
• When the KPI should turn Red

There are no hard and fast rules to the meanings attributed to each of the coloured areas but generally they tend to be:

• Green – an acceptable result, we are on target
• Amber – there may be a problem, we should investigate
• Red – an unacceptable result, there is a problem that needs rectification

By using an example ( for instance: Day to fulfill an order) we can illustrate as follows:

• Green – 60 days or less
• Red – 80 days or more

When these thresholds are entered into a performance management system like QuickScore the result might look like this:

Why Red Amber Green - A

 

As can be seen, by setting threshold values the viewer can instantly and very graphically see the current situation and equally importantly the history leading to this point.

Out of the numerous threshold models there are two more that need to be looked at, the first is a simple extension of the RAG model, the second a variant that accommodates measures that are not linear in nature.

Often there is a need to get a better understanding of an ‘over-achieved’ status. This is particularly true in the area of sales where bonuses may be based on not only achieving a target but over-achieving it as well. In this case an extension of the RAG model can be used; the Red Amber Green Blue variant. Here, it is normal to set five thresholds:

• The lowest acceptable result
• When the KPI should turn Red
• When the KPI should turn Green
• When the KPI should turn Blue
• The highest acceptable (or capped) result

This way an over-achieved status can be monitored and managed, most system will also put a ‘cap’ on the highest acceptable result. Again in sales this may be desirable to avoid run-away bonus schemes. The result might look like this:

Why red amber green RAG B

 

The third threshold example is the ‘stabilise’ KPI. Occasionally KPIs are deemed unacceptable if the result is either too high or too low. A good example is a training budget. In training we want to spend to the budget but not exceed or go below the budget. In this case we define the ‘best’ result and then determine acceptable and non-acceptable results below and above best. Using the example of a training budget, the result might look like this:

Why red amber green RAG C

 

One of the by-products of defining thresholds is the ability to turn the KPI into a relative score. You may have noticed in the gauges above together with the actual value and the threshold values a ‘score’ has appeared. In the cases above, a calculation has been applied all of the actual values to create a normalised score out of ten. This allows the performance management system to compare ‘apples’ with ‘apples’ and roll-up the scores to higher levels including objectives and perspectives, departments and divisions and finally to an overall company score. (This becomes a very powerful feature when building company dashboards).

For more information on how to develop meaningful key performance indicators, why not take a look at our White Paper