Management As A Science

admin
Management As A Science

MAAS – MANAGEMENT AS A SCIENCE

Let’s emphasize the message in a flashy way, with a new letterword: MAAS, not “As A Service”, but yet “As a Science”. More particularly “Management As A Science”: MAAS! Another plea[1] for a more scientific approach of doing business and management. 2 examples to illustrate how this can add value in real life.

The first case is a concise summary of an article[2] of Nobel prize-winner Daniel Kahneman. The second case is a description of a rather unknown methodology to come up with good quality estimations for, amongst others, digitazation projects.

Lacking time? please feel free to skip one of both!

VENTURE CAPITAL : HOW TO STRENGHTEN THE INVESTMENT DECISIONS!

Many companies have processes in place to make decision more objective and fact-based (ie procurement as a mature process). For the important, strategic decision though this is quite often not the case. Saying this, it opens for that kind of decisions (acquisition? Investment, yes or no? Recruitment of a new director? New market, product? …) all pitfalls of the individual and groupthinking.

Daniel Kahneman is probably the most cited living Nobel prize-winner of the moment. His work around human beings’ thinking have changed, without a doubt, the insights on the mechanisms of our thinking and predominantly on the error-prone character of it. His book from 2011 ‘Thinking fast and slow’explains and illustrates thoroughly the baked in errors in our thinking and hence decision-making.

A good illustration of this that has been documented is described in ‘the body of research on the job interview’ that contains a wealth of information[3] on the topic of recruitment methodologies. A vast amount of evidence discloses the weaknesses of guts-based selection procedures. What is hardly appreciated in boardrooms is that the same problem of the ‘mental model’[4] is present in strategic decision-making.

Indeed, this is also true for senior management or the investment board of a venture capitalist. It is worth it to analyse the impact on  investment decisions of the structural mistakes, Kahneman has extensively documented in our thinking and decision-making. His article we summarize here, elaborates the methodology called MAP, Mediating Assessment Protocol. He explains the application of MAP in a venture capital firm: a strengthening of existing decision policies. “The purpose of MAP is to delay intuition. The structured process calls for analysing a decision based on six to seven previously chosen attributes, discussing each of them separately and assigning them a relative percentile score, and finally, using those scores to make a holistic judgement” (quote from article” Leadership Analysis: Nobel Prize-winning psychologist to CEOs; Don’t be so quick to go with your gut”, The Washington Post, March 2019). In fact this pushes aside the holistic discussion, it is true, based on quantitative information of several attributes/parameters (market, management, product, operations, supply chain …) and lead to a discussion per attribute (most probably based on the same information, if professionally documented) to have only in second instance (delayed) the holistic discussion on the file, on an investment, on an acquisition or any other decision, but now based on the separate, mutually independent scores of the evaluations per attribute. Through the attribute per attribute approach some of the ‘structural erros’, (biases, baked in in our thinking) of our thinking are blocked or mitigated. Kahneman and colleagues explain how this Mediating Assessment Protocol helps the decision committee to avoid the structural errors we’re all making in our thinking[5] and hence in our decision-making, even if we discuss the potential investment, based on the same information but in an holistic way, without the intermediate step. Knowing and accepting the individual shortcomings in our rationales is not enough, but the methodology, developed by Kahneman will redirect and improve the strategic decision-making.

For more detailed information on MAP, I can refer to the article, cited above.

PROJECTPLANNING, THE EVERLASTING UNDERESTIMATION

Every professional, from close or far experienced with change projects will confirm the biggest challenge: planning.  Why does deployment of a change require so often more time (and money) than planned? Underestimation is a latent hurdle to step up to successful projects, projects within the limits of the 3 parameters: time, budget and quality. And here he is again, Daniel Kahneman in another publication[6], with colleagues and the planning fallacy as a phenomenon of structural underestimation. Estimations and according plans undermined by the optimism bias.

Other psychological and social elements are playing a role: developers, looking for challenging coding but annoyed by the questions to estimate the workload (‘I”ll tell you when I’m finished’) and therefore growling confirming what the project manager is whispering in his ears. The young ambitious project manager willling to please his sponsor by a self-confident ‘yes, we can’ and underestimating potential efforts, complexity and risks of the project. The sponsor, with a lack of experience with technology pushing for ambitious targets (‘My 13 years old sun, develops an app in less than a day’). Biased estimations, leading to bad investments, waste of money.

Kahneman shows us in a scientific way the errors we make estimating future efforts. The Delphi-protocol can be part of the solution to mitigate our failing thinking and decision-making. The methodology was developed by the American army in the second world war.  The name “Delphi” derives from the well-known Oracle of Delphi. The method is based on the assumption that group judgments are more valid than individual judgments.  Let’s have a closer look to a digitization project and how the estimation challenge can be faced. An iterative process:

  • Put around the table a number of experienced people, experienced with similar projects, but not necessarily on the execution side. Sponsors are also welcome. And not necessarily with experience on the complete track of a project. I.e. the system engineer with responsibility for the infrastructure bit (hardware, system software and middleware).
  • Split up the project in a limited number of phases (you can use standard phases of a project methodology) or releases (requirements). Too much detail on the split is not required.
  • Everyone around the table documents his or her estimations for the different phases, accompanied with assumptions, supporting, justifying the estimations.
  • All estimations are disclosed to the group. Next step is important: everybody explains the rationale behind the estimations, by explaining the assumptions. As you will see, most of the differences are not simply because of different views on the workload, but because of different assumptions: other work items in a phase, major requirements ex- or included, definition of phases, level of required quality, sourcing, support …

As a group, you decide together to accept or refuse assumptions in order to evolve progressively to a shared view.

  • After that, everybody adapts his/her assumptions as agreed and reviews accordingly his or her estimations, based on the new set of shared assumptions. New assumptions may be formulated (and will be discussed again as next step, together with the estimations).
  • This iteration is repeated 2, 3, 4 times. A good moment to stop is the moment all assumptions made are more or less reconciled, harmonized and no new assumptions are rising. At that moment the majority will have a consensus on the estimations. Different views on that moment are pure discussions on workload (‘the development of a new workflow takes 20 days or 25 days?’) and a consensus (avoid decisions through authority) is not too difficult.

A bit of training on the application of the Delphi protocol is required, but the results are great[7]. The planning fallacy , optimism bias or errors through groupthinking are tackled.  Science disclosed errors in our thinking and decision-making. A structured methodology, Delphi, applied in a pragmatic but professional way (an average of wild guess isn’t ‘a delphi’),  improves the result of the work delivered. The results are worth the (rather limited ) effort.

 

[1] Interested in more on Management and Science: https://kobusconsulting.be/en/management-ondernemen-en

[2] “A Structured Approach to Strategic Decisions” ( MITSloan Management  Review, Daniel Kahneman, Dan Lovallo, and Olivier Sibony, March 04 2019)

[3] Julia Levashina, Kent State University, Christopher J. Hartwell, Purdue University, Frederick P. Morgeson Michigan State University, Michael A. Campion, Purdue University, “The structured employment interview: narrative and quantitative review of the research literature”, Personnel Psychology 2014, 67, 241-293

[4] Mental model: notion used by Kahneman and defined as an impression of a complex situation, often less nuanced and more coherent than the reality it represents.

[5] Daniel KAHNEMAN, Thinking Fast and Slow, Farrar, Straus and Giroux; 1 edition (October 25, 2011), ISBN-10: 9780374275631, ISBN-13: 978-0374275631, ASIN: 0374275637

[6] Buehler, Roger; Griffin, Dale, & Ross, Michael (2002). “Inside the planning fallacy: The causes and consequences of optimistic time predictions”. In Thomas Gilovich, Dale Griffin, & Daniel Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment, pp. 250–270. Cambridge, UK: Cambridge University Press.

[7] A very limited analysis, without scientific ambition but interesting though. A portfolio of over 3.000 mandays with Delphi-estmations’s led to an underestimation of 7%. A second portfolio  of almost 3.000 mandays with traditional estimations showed an underestimation of 29%. Both portfolio’s were aggregated from a long lists of finished projects, more or less successfully from a quality view-point.