/* Code added to allow category posts to be displayed as masony format and get pagination correct. See https://www.elegantthemes.com/blog/divi-resources/how-to-give-your-divi-archive-pages-a-masonry-layout */
UK +44 1780 784887 info18@excitant.co.uk
Select Page

Why use a KPI?  There is a lot of discussion about which particular measure, indicator or KPI to use.  There is far less discussion about why use a KPI in the first place.  Actually, the reason is often a heuristic.  Let me explain….

To solve difficult questions, we often look for a simpler question to answer.

I am still reading the excellent book, Thinking Fast and Slow by Daniel Kahneman and thinking (slowly) through the implications for how we manage strategy, people and performance in today’s organisations. 

I was reading about how our System 1 (thinking fast) and our System 2 (thinking slow) interact when faced with a difficult problem or question.  In some cases system 1 will find a simpler question to answer that acts as a substitute for the real question that it is facing.  Sometimes, this substitution of one question with another, can occur extremely quickly (the System 1 shotgun effect). 

Often, this substitution occurs so quickly that we don’t even notice that we have done it. I particularly liked the line, describing the effect:

“The mental shotgun makes it easy to generate quick answers to difficult questions, without imposing much hard work on your lazy system 2”.

Daniel Kahneman refers to these two separate and different questions as “The target question” and “The heuristic question”.  The target question being the one that really should be addressed.  The  heuristic question is the one that is substituted as a simpler question in order to answer the more complex target question. 

[Note, a heuristic is a simple procedure that helps to find a simpler, but adequate, though often imperfect, answer to a difficult question.  For more information, see ‘ Kahneman, D, 2012, Thinking Fast and Slow, pp97-98]

Now substituting one problem for another can be a good strategy when faced with difficult problems.  Kahneman, cites several examples such as when faced with the question, for instance, “How happy are you with your life?”, we can substitute that for “What mood are you in right now?”.  However, as he also cites examples, it can lead to problems.

Is Asking for KPIs a substitution heuristic?

Whilst reading this I was thinking about how so many times I hear people simply say, “I want some KPIs”, or ask, “What is a KPI or measure for this situation”. I usually think, “Why use a KPI?” but it is east to simply answer the question by providing some KPIs.

The question occurred to me, “Is this a substitution heuristic?”   If it is a ‘substitution heuristic’ then what is this statement , “Please go and find me some KPIs” a substitute for?  What are the potential target questions?

Is “How do I measure?” is a simplistic substitution for, the far more subtle question, “How do I manage?”. 

We know that management is far more complex and involved than mere measurement.

Immediately (thinking fast – sorry) it occurred to me that this is a substitute for the question “How do I manage and control this complex situation?”.  Now in many cases management is complex and requires thought, a deep understanding of the situation and people involved, a clear idea of what you want to achieve, clarity about the relationships and some subtle person skills.

What is worrying is where the “Find me some KPIs” question is a heuristic substitute for the clearly more complex and difficult question, “I find this area difficult to manage, so if I simply go and get some KPIs for this area, life will be simpler and my problem will be solved.”

“Find me some KPIs” can be a heuristic substitute for the clearly more complex and difficult issue, “I find this area difficult to manage!”. So if I simply go and get some KPIs for this area, life will be simpler and my problem will be solved.”

Finding a KPI has become a substitute for thinking through how to manage the people, area or situation.  Frankly, it is a very weak and approximate heuristic. Hence the question, Why use a KPI?

The word simple function of “Measure” is being substituted for the more complex idea of “Control” or “Manage”.  These are clearly not the same.  Why use a KPI?  Is it the best way, or only way, to achieve control and manage the area?

Why do you want a KPI?  Why use a KPI?

A way to run into trouble at this point is to be too blunt.  It is usually inappropriate to ask the direct question “Why do you want some KPIs?”.  It is very likely to be met with the equally blunt response, “Don’t be so cheeky, just go away and get me some KPIs!”. 

You can ask, ” Why do you want these KPIs?”  The more subtle question is “How will these KPIs be used?”

However there is a more subtle way to address the underlying substitution heuristic.  That is to ask the more subtle and interesting question, “How will these KPIs be used?”

The question asking how the KPIs will be used, is (more politely) starting a conversation about the process of thinking that went on between the two questions:

  1. How best to manage this area;  and
  2. Which KPI to use.

It is asking the question about the intention and the management approach that is behind the thinking that a KPI would be useful.  Effectively you are asking, useful for what?

A leading KPI for what purpose?

What about “I want a leading indicator”?

The simplistic substitution heuristic used when discussing leading indicators is often something like, “I want to know in advance that things are going right, or are going wrong.”  Even this is not a simple question. 

“I want a leading indicator” usually means, “I want advance notice of things going wrong”

Not so long ago I was asked to help create some “leading indicators” for a particular industry, and companies within it.  However, what was not clear is what aspects of their industry and the various company’s situations that these indicators were to be leading indicators of.  I wrote an article about some of the differing types and classes of leading indicators that serve quite different purposes in different situations.  To find out more, have a read of  Making sense of Leading and lagging indicators and “Ten different types of leading indicator”.

You need to be especially clear, when dealing with leading indicators, which particular flavour of substitution heuristic is being referred to.

KPIs to be useful to whom?

You can also ask, useful for whom?  This is a really important question because it goes to the area of control vs learning.  Is this for that team to better understand what they need to achieve?  is it for outside control.  Is it or managers or some other team (perhaps finance) to achieve their objective? It is for some external body, perhaps customers, suppliers or auditors?   

“Who is going to use this measure? And in what way?

It is vital to understand how it will be used by any of these groups, if at all.

Unpicking what the KPI is trying to communicate.

It is very common in measure methods and KPI approaches, to move back from the KPI to ask what is the objective and what are we trying to achieve in this area.  We insist that teams develop their objectives, before they touch any measures. 

“Develop your objectives, before you touch any measures.  Work out what the measure is trying to communicate.”

In our Strategic Balanced Scorecard approach, this is simply stated as “Objectives before measures”.  The process of developing a rich understanding of the objective and its characteristics, also helps immensely in providing other information that will accompany any measure to further explain where it came from.  The objective explains why you use the particular KPI.

Having the objective, before the measure helps where a direct measure is not available and an indicator or temporary surrogate can be usefully used.  (See the difference between measures and indicators).

The false substitution here is “Knowing this measure” is not the same as “Knowing what is important and the right way to achieve it”

What behaviours will this KPI encourage?

We know that all measurement is really a part of the wider issue of management: Not just performance management, but of the management of a whole system. 

“Check what behaviours any measure might encourage.”

We all can cite many examples of unintended consequences of poorly designed and implemented KPIs distorting behaviour.

For these reasons, we always use the test question “What behaviours wil this encourage?”.  Our clients find this useful in thinking through the potential for unintended consequences of measures and KPIs.

The deeper question “How does the KPI fit into our approach to management”

So when faced with, “Why use a KPI?” it is useful to thinking about who will use the KPI, how it will be used and what potential behaviours it might encourage.

However there is a wider question to do with how we manage.  We also need to ask, “How does the KPI fits into the philosophy of management and our management system?”

The wider question: “How does the this KPI fits into our wider philosophy of management and our broader management system?”

So often we find that the behavioural and cultural implications of simplistically applied measures is that they clash with other aspects of the way we manage.  For example, we have seen an attempt to introduce a new way of funding changes stifled by the constraints of the financial system and management accounting.  Moving to continuous strategy requires rolling budgets and the flexibility that annual goal setting and appraisals rarely provide.

Conclusion

So, beware the KPI substitution heuristic.

“Our organisations are rich, complex social systems: they are not simplistic mechanisms capable of being managed merely by measures and targets.”

Find ways, politely, to unpick the substitutions that are occurring within the idea of “get me some KPIs”.  Ask why use a KPI?  Aks how it will it be used?  Do make sure that in addressing the simpler problem, (I just need to find some measures for this situation), you are not creating problems caused by the simplification of the problem.

After all, our organisations are rich, complex social systems and our performance management system are tools of conversation and thinking: they are not simplistic mechanisms capable of being managed merely by measures and targets.