• Published: Dec 12, 2023
  • share on linkedin
  • share article

First published in Board Works #5, 2010 

In the boardroom decision-making environment, whether we challenge or accept other people's conclusions and assumptions, we need to be confident that their reasoning—and ours—is based firmly on the best information we can get.  

The challenge is to make sure decisions and subsequent actions are as transparent and grounded as possible. The type of dialogue that will achieve this faces many impediments. Not least, time at board meetings is invariably in short supply. We are always under pressure to deal with matters here and now.  

So we often find ourselves processing what are little more than peoples’ self-generated beliefs. When we adopt and act on flawed beliefs without adequate testing, we often ‘get it wrong’. We leap to counterproductive conclusions producing poor decisions and unproductive conflict. 

A very useful conceptual model can help understand why—and how naturally this happens—and how to avoid this situation. It was developed by organisational psychologist Chris Argyris [1] and is referred to as the ‘Ladder of Inference’.  

Here is a brief scenario to demonstrate how the ladder works and how quickly we climb it: 

I am chairing the board. We are discussing an important matter I have brought to the board via a thorough and well-argued board paper I have worked on with the Chief Executive. This concerns an issue about which I feel very strongly, and I consider that the board can, and should, decide on it today.  

The directors came to the discussion well prepared, and all seemed to have been engaged and alert, except for Justin. He has appeared somewhat distracted throughout the discussion. It looks like he has been preoccupied doing his emails on his iPhone under the table. Unlike the other directors, he hasn’t taken the opportunity to share his thoughts except for one question that was clearly answered in the paper.  

The others seem to have got their heads around the analysis of the situation and the solution I have proposed. I am thinking that we have reached a consensus and can make a decision. Right at that moment Justin interrupts my summing up and proposes that the issue be deferred pending a full, independent report.  

I am angry that, having not really participated, Justin is now cutting across the feeling of the group. I conclude that the only reason Justin wants to kick this for touch is because it concerns something that is obviously important to me.  

Come to think of it he always opposes things that I promote to the board. This is another sign of the power trip he is on. He obviously wants me off the board and to become chair himself. I will make sure he doesn’t get anything he wants on the agenda in future, and I will start lobbying the group that nominated him onto the board to replace him.  

What has happened here? 

  • I started with the observable data: Justin’s less active participation in the discussion and his proposal to defer.  
  • I selected some details about Justin’s behaviour—for example, his apparently greater interest in his iPhone. 
  • I added some meanings of my own, based on the board’s norms (eg, thorough board meeting preparation and Justin’s apparent lack of it). 
  • I assumed that Justin is trying to defer the proposal because it is important to me.  
  • I concluded that Justin is out to get my job. 
  • I have reached the top of the Inference Ladder and I am plotting against him. 

I have moved up the steps in the ladder so fast I have not been aware of it. The only verifiable aspect (observable by others) was on the lowest rung. The need to act to get rid of Justin (at the top) is the only thing I have a residual and comparatively conscious sense of myself. The rest of it has been a series of leaps up the middle rungs of the ladder. These occurred in my head and so were not visible to others (or even myself!). So, they could not be questioned or discussed (tested) and were very abstract. These steps are sometimes referred to as ‘leaps of abstraction’. 

The more I believe that Justin is malevolent towards me, the more I will look for evidence of that behaviour in the future. The reflexive loop in Justin’s mind, as he reacts to my increasingly antagonistic behaviour towards him, is likely to induce him to start a rapid ascent of his own ladder! Before very long we might find ourselves becoming very confrontational. 

What are some alternative explanations? Justin may have had some serious issues going on back in his own business, or even in his family, to explain his apparent distraction with email. His comparatively low level of engagement in the discussion may have reflected his satisfaction with points his colleagues were making. His proposed deferral may have been because he felt that the board (and me in particular) were too close to the issue, and we needed an independent assessment to verify (or shift) our own thinking.  

We can’t know any of this without testing our assumptions and conclusions. Unfortunately, for some of these it may be particularly difficult. I may be able to take Justin aside at a break and, sympathetically, ask him if there is anything going on that is distracting him. It is far more difficult to find out if he is after my head! It is far easier to have these types of conversation while still on the lower rungs of the ladder! 

Jumping to conclusions in this way is common but understanding that this type of process takes place is a good start in reducing the problems that can arise. You can also do other things to manage the risk: 

  1. Make your thinking process more visible to others. Try to be more active in explaining your assumptions, interpretations and conclusions.

  2. Invite your colleagues to test your assumptions and conclusions. Work with them, encouraging them to help you think things through

  3. Help them make their own thought processes visible as well. Use open and non-judgmental (‘learning’) questions.  

  4. Encourage the board—as a matter of board norms—to value the existence of differences among directors and to take the opportunity to explore these transparently. Don't agree to disagree too soon. Try to surface hidden or unspoken assumptions and conclusions before progressing too far up the ladder. 

Next time you have an emotional reaction to something like the Justin situation, try to identify what triggered this feeling. What happened? What data did you select? What filters did you use to interpret the information? What were your assumptions and conclusions? What were the consequences (or potential consequences) in acting from the top of the Inference Ladder? 



[1] Peter Senge and others (1994) The Fifth Discipline Fieldbook: Strategies and Tools for Building a Learning Organization. London, Nicholas Brearley Publishing