Sergio Caredda
In a fascinating article on HBR by the title “The Dangers of Category Thinking”, Bart de Langhe and Philip Fernbach, analyse the critical issues related to thinking in categories in several domains. Humans tend to simplify reality by splitting items into groups, and in its nature, it derives from our capability to survive.
Your mind is a categorization machine, busy all the time taking in voluminous amounts of messy data and then simplifying and structuring it so that you can make sense of the world. This is one of the mind’s most important capabilities; it’s incredibly valuable to be able to tell at a glance whether something is a snake or a stick.Bart de Langhe and Philip Fernbach, The Dangers of Category Thinking, HBR Fall 2019
Derek Cabrera further states that, in many ways, categories are useful structures for organising information. “There are many ways in which thinking is an act of classification, so categories are embedded in cognition itself. Because they help us situate and structure information, using categories can make us feel like we are building knowledge and progressing in our thinking“.
However, for categorisation to have value, two things must be true: First, it must be valid. Second, it must be useful.
Seems easy, correct? But the reality is that in business we often rely on categories that are either not valid, or not useful or even a combination of the two. Wich can lead to errors in decision making.
When we think about the validity, we need to consider the extent that the categories reflect reality. We have already mentioned in the book review “The End of Average” how much the problem of using mean data to create groups, ends up creating things that do not exist in reality. A vital example of this is Customer Profiles, often statistical artefacts that are invalid. Plus very often we fall in love with the concept of the category itself. But are they beneficial? They cite, for example, the MBTI personality test, which despite its widespread usage, fails to have practical usefulness in predicting potential at work, for example.
Naomi Stanford has created a list of other categorisation tools used in a business context that in her opinion, provide “nonsense” to the world, thus would fail the test of validity and usefulness. From grading systems to competency models, from matrixes to RACI charts, all these tools “imposes unhelpful, artificial boundaries that can hamper considered ‘design’“. With my ols consulting hat on, I would add any 2×2 or 3×3 matrix that seems to dominate our way of thinking. And I would add as well OrgCharts, as these often create artificial categorisations when visually splitting an organisation.
Again Cabrera points out to one area where categorisation failed miserably: Web Directories. We all remember how Yahoo! worded. It organised things into categorical directories, a little bit like books are arranged in our libraries. The results were that when you searched for something, you never really found what you needed. Instead, drilling down into part-whole structure led to a terminal end (a category cul de sac) which was far afield from where you started and almost always irrelevant.
The Map is not the Territory
This sentence is a famous quote by polish philosopher Alfred Korzybski and calls precisely on the issue of separating the concept of reality from its representations.
De Langhe and Fernbach identify four dangers in the way Category thinking can be dangerous.
- Compression. Thinking in terms of categories can push you to compress the members of a category, treating them as if they were more alike than they are in reality. Target Customers in marketing are a perfect example of this, as they blind us on the variety that exists into each cluster we create.
- Amplification. In this case, we can exaggerate the differences between categories and segments. One example is when we categorise people by a political view, or by social belonging: we tend to polarise the perception of the groups.
- Discrimination. This happens when you favour specific categories over others. It often happens in marketing with targeting campaigns. Even though consistent studies have shown that targeting broadly yields better ROI than targeting narrowly, over-targeting has become an issue, mainly also thanks to the spreading of digital advertising, that allows to real specify narrow criteria for targeting.
- Fossilisation. This happens when you treat categories as static, keeping them “as-is” for a long time. It affects a lot innovation, and in many cases organisations that were not able to exploit innovations, did so because of old and fossilised category thinking. Think about the famous quote by Ken Olson, CEO of DEC, who stated: “There is no reason for any individual to have a computer in his home.” For sure, there where marketing studies and clustering of buyers profiles behind this assumption. But the reality was different.
What can we do to to escape Categorical Thinking?
Tha author continue by presenting a few solutions to limit the dangers of Categorical Thinking.
- Increase Awareness
- Analyze data continuously
- Audit decision criteria
- Run “defossilization” meetings.
[…] today, as the data revolution progresses, a key to success will be learning to mitigate the consequences of categorical thinking.Bart de Langhe and Philip Fernbach, The Dangers of Category Thinking, HBR Fall 2019
I would add another critical element to the four above Systemic Thinking. Born as a consequence to the General Systems Theory initially developed by Ludwig von Bertalanffy, has been developed as a set of tools and methodologies focused on looking at interdependencies of the parts of a system, rather than on the individual components.
Thinking in terms of systems can help us overcome the limits of categorical splits. And I genuinely believe that Systems Thinking needs to be seen not just a toolkit or methodology, but as a skill to be developed and built for successful leadership.
And you? How many categorisations do you think exist that can lead to wrong decision making?
Comments and Feedbacks
More posts like this
