How (ir)rational are we in financial controlling?

Widespread focus on methods and tools

Financial Controlling as a function usually has the task of ensuring rationality in the company - but how rational is controlling itself and what does it entail?

Expectations of today's controlling are high - a wide variety of roles, entrepreneurial dynamics and state-of-the-art tools keep managers and controlling specialists busy.

The focus on one core task - preparing and making decisions - quickly recedes into the background. However, it is precisely this area that is particularly critical.


In 2020, a study by the University of Giessen addressed the question of what role the human factor plays in the decision-making process. 90% of the respondents (primarily CFOs, heads of controlling and people in controlling) stated that unconscious biases can have negative consequences for their company. In addition, three out of four respondents attributed wrong decisions in the past at least partly to biases.


There seems to be some awareness that human factors can play a negative role - but to what exactly? Who is primarily affected by biases?

Irrationality in financial controlling

Homo economicus is an assumption that runs through economic theory. Rational "Econs" make perfectly rational decisions and optimise for the best of the respective company. Humans, on the other hand, believe in their ability to influence things more than they are actually able to. They confuse easily retrievable information with important information, tend to be overly optimistic and increasingly perceive information that confirms their prevailing opinion. And much more.


Do controllers belong to Richard Thaler's "Econs" or to the "Humans"?

Freedom from prejudice, objectivity and neutrality are attributes that are important for the profession and probably also belong to the role's own image.

Can we assume a fundamentally stronger rationality in controlling compared to other functions?




The human brain fundamentally works selectively and with a wide variety of thought shortcuts. Only a small fraction of the information arriving every second ends up in consciousness.

A Duke University study examined CFOs' ability to predict the market (S&P stock index). More than 11,000 forecasts showed a correlation of less than zero - when forecasting a market decline, the correlation increased and vice versa. At the same time, the forecasts were queried with a confidence interval of 80%. This means that while the CFOs in this study were confident in their ability to forecast the market, on average they did not have this ability.

Can we conclude from this example of overconfidence bias that Controlling & Finance may be more affected by biases than others?




We can assume that thinking biases and faulty conclusions occur in this function as much as in all others. Unconscious biases have a universal effect.

However, the job profile in general and the personal identity in particular can lead to a less open-minded attitude towards one's own biases.

Is digitalisation & AI the solution?

The demand on controlling to ensure rationality in the company means that the function can fall back on a broad toolbox. The methodological competence and the possibility to (co-)design processes in the company already represent effective de-biasing potentials - biases are also anchored in processes, structures and procedures in companies and therefore affect more than the individual level.


Controlling is therefore in a fundamentally good position to have a positive effect on the basis of corresponding knowledge about unconscious biases in general and the concrete effects in particular (therefore on the acting persons or processes).


What potential is there in increased digitalisation or artificial intelligence?

In principle, these represent further methodological possibilities for making assessment and decision-making processes more rational. However, two important points need to be considered here:

  • Algorithms are developed or (in the case of self-learning systems) influenced by humans - as cases of discriminatory recruitment software show, these are not bias-free by definition.
  • Other systems generally support decision-making processes, but do not "dehumanise" them completely - apart from the question of whether this would be desirable, this means that biases still have an effect, only possibly differently.


The ICV (German for: International Controller Association) states that controlling work requires not only methodology but also knowledge of behaviour. Knowledge about one's own behaviour as well as that of others makes the work effective - this also includes acquiring competencies about the factors that negatively influence this behaviour.

Not much happens

The aforementioned survey by the University of Giessen also asked how many companies have implemented systematic de-biasing - 12%!

Nine out of ten companies recognise the problem, only one out of ten does something about it.


De-biasing is not trivial - in addition to focusing on the individual context, it also requires the willingness to question or change one's own thought patterns and ways of acting. The protection of one's own identity sometimes speaks against it.


Nevertheless: there is undoubtedly a lot of potential - not least because it is a challenging topic that not every organisation is ready to master.

What can we take away for financial controlling?

The broad set of competences that are needed to perform in a modern controlling role means that:

  • constant learning and development is necessary
  • knowledge on methods, tools and processes is not enough
  • there is potential in addressing unconscious biases as others are not focussing on it

There is potential - personally to be accepted as a valuable business partner and for the company due to more effective decision support.

Mind your business uses a holistic perspective on controlling & finance, putting de-biasing as a key contextual element next to digitalisation and sustainability.

Take advantage of this opportunity!