Asset 1

Where do all these models and frameworks come from?

June 30, 2024

So, where do they come from? The source, whether we’re talking about OKRs, Scrum, or the latest must-follow leadership framework, boils down to a Venn diagram of science on one side, and fiction on the other.

Circle 1: Science

Science, or at least the scientific method, is often the way that we think that methods are created. That through research, analysis, and refinement, effective options are generated, and indeed, there are plenty of examples through history.

The guild-systems in our medieval past, where luminaries like Johann Joachim Becher helped to create standardised structures and practices that allowed things like “vast quantities of standardized muskets, uniforms and boots [to be] turned out of countless artisan shops across Europe.”[1] The activity studies of Frederick Windslow Taylor, the American Engineer, who – by analysing tasks and standardising tools, created efficiencies. Efficiencies that were carried into late 20th Century thinking by other titans of industry, like Sloan at General Motors. Or more recently, Google’s experiments into the need for managers, and the constant plan-do-study-act of organisations like Spotify and Buurtzorg to develop and refine their working practices.

Alongside, we have the many psychologists and social scientists conducting experiments into what make us tick. Consider Kahneman and Tversky’s work on thinking systems and biases, Edmonson’s work on psychological safety, or Deci’s work motivation and reward. The list goes on.

Circle 2: Fiction

Fiction, which includes pseudo-science and cause-effect confusion, makes up the other half –well, the majority, really – of the Venn diagram. To break that down, let’s look at some examples.

Pseudo-science.

There is a big difference between the natural sciences where proof is possible and demanded, and social sciences, where it is neither. Say we are conducting an experiment in a laboratory. We have plenty of options to monitor and manage our inputs and process, dealing with confounds by using controls: One group takes the new drug, another takes a placebo, different alloys are subjected to identical tests for stress and durability, and so on. More, a paper that does not adhere to this kind of rigour with be pulled apart by peer review or simply not get published.

For social science, it is much less like this. Practically, control is affected by numerous confounds (are two groups and the work that they are doing really the same?), and a recently published exchange of letters about authentic leadership illustrates perfectly.[2] There are difficulties in describing what authentic leadership actually is, and even with a definition, how can we know that it was that-one-thing that helped? As Gardner et al. go on to describe:

“We have yet to see proof for (or refutations of) Cohen, March and Olsen’s garbage can model, Maslow’s hierarchy of needs, Freud’s structure of the mind or Smith’s invisible hand.”

That’s interesting. How often has a leadership course introduced Abraham Maslow’s hierarchy of needs by saying: ‘remember none of this is proven, and it might not be true, or relevant.’  Personally, I think we should be even more sceptical of the ones that rhyme. Forming-storming-norming-performing, anyone? People know those words, but less that the original research was completed on a small group by Tuckman at the US Naval Medical Research Institute in Bethesda in 1963-65. In the context of our increasingly online lives, this has little or no relevance – or at least, no deeper review to confirm its validity.

I am certainly not saying that social-science research is valueless, rather that I agree with Garder et al. when they invite more precision.

Cause-effect confusion.

Google is well known for making use of OKRs, and I can argue it is through John Doerr – investor in Google and many other tech companies – that their popularity has flourished. I mention Doerr because the subtitle to his book reads: “How Google, Bono, and the Gates Foundation Rock the World with OKRs.” Is it true that Google have used OKRs? Yes. Is it true that Google’s success can be attributed to their use of OKRs? No. Or at very least, it isn’t knowable. As above, we do not or cannot have a control in the experiment. That is, we don’t have a Google that didn’t use OKRs to compare and assess their effect. It could be that they did, but it could also be that they would have been more successful without!

A similar example relates to the leader being asked what made them successful. ‘I was authentic, gave people lots of space, and expected high standards’ or some such. How do you know a) that you did, and b) that it was this that made you successful? It’s a nice sound bite but has no veracity.

Fiction.

Finally, and briefly, we have the things that are just made up. Things that have no link to research or experience or are cobbled together to sound nice. We can all imagine those things, so I won’t dwell, but I will suggest you read one of my favourite ever published papers.

Written by Mats Alvesson (who was part of the team challenging the proof of authentic leadership), it provides a step-by-step guide for the up-and-coming leadership studies professional, wanting to generate their own leadership framework.[3] Including such concepts as ‘Lump some variables together and (re)produce a ”style” and ‘Take artificial data seriously’, it gives you everything you need. Seriously, however, Alvesson also proposes what you should do if your target is not merely a marketing exercise, and therefore how to further the field of study.


[1] Farr. J, Artisans in Europe, 1300-1914, (2000), Cambridge University Press. ISBN 978-0-5214-2934-4

[2] W.L. Gardner, E.P. Karam,M. Alvesson, et al., Authentic leadership theory: The case for and against, The Leadership Quarterly, https://doi.org/10.1016/j.leaqua.2021.101495

[3] Alvesson, M., 2020. Upbeat leadership: A recipe for – or against – “successful” leadership studies. The Leadership Quarterly 31, 101439. https://doi.org/10.1016/j.leaqua.2020.101439