One of my favourite tools for promoting productive conversations in teams is the Ladder of Inference.

I first came across it more than twenty years ago when we were trying to change our ways of working in a huge global reorganisation and transformation.

The phrase was coined and the concept developed by the great Chris Argyris. It explains why we are all capable of ‘jumping to conclusions’, rightly or wrongly, and gives us a neutral technique to explain our own thinking or, more likely, to get others to explain their thinking to us.

The Ladder of Inference describes how you move from some observable data (something someone says or does for example) through to making a conclusion and maybe even taking action based on that conclusion.

An example. I am giving a presentation. But one of my peers hasn’t said a word and she normally does. She must be bored with what I’m saying. She thinks it’s not important. And she probably thinks I am incompetent.

But do I know this for sure?

We all jump up the Ladder. We select the observations we will treat as important and ignore the rest. We add meaning and draw our conclusions. Our thinking is effortless, fast and works routinely. Our conclusions appear obvious and we rarely think about the steps we’ve taken to get there. And our beliefs, experience, assumptions and values influence what data we select, the meanings we add, and what conclusions we draw.

It is not always bad and the skill is essential to everyday life. For example if you are crossing the road and a car suddenly comes directly towards you at speed, you want to get immediately to the top of your Ladder, conclude you might be in danger and take action to get out of the way!

However it can get us into trouble. We jump up the Ladder without stopping to reflect. People with the same data can often reach different conclusions. But they see their conclusions as obvious and do not see the need to explain the steps that got them there.

And when people have the same data but draw different conclusions they can argue (and perhaps shout) from the tops of their respective Ladders. Which I have always found a wonderful picture.

All of our thinking is therefore vulnerable to error. And if we don’t test our thinking openly, the errors remain. Which can have an important impact on the quality of the thinking, discussion and the decisions made as a team or organization.

The seven steps on the ladder from top down (although obviously you start climbing from the bottom of the ladder not at the top!) are:

  1. I take actions based on my beliefs
  2. I adopt beliefs about the world
  3. I draw conclusions
  4. I make assumptions based on the meanings I added
  5. I add meanings (cultural and personal)
  6. I select data from what I observe
  7. Observable data and experiences (as if recorded in a video)

And all of the steps of the Ladder take place in my head, invisibly. Other people only see the observable data and my own decision to take action. The rest is unseen and untested.

But sharing the other rungs, using the Ladder of Inference, can improve communication and productive conversations in three main ways:

  • Reflection: becoming more aware of your own thinking and reasoning
  • Advocacy: making your thinking and reasoning more visible to others
  • Inquiry: inquiring into other’s thinking and reasoning

If you want to read some more on the Ladder of Inference and other techniques for dealing with mental models then you could do worse than look at Peter Senge’s wonderful work ‘The Fifth Discipline Fieldbook”, which is where the seven steps of the Ladder and the three bullets above come from.

Another example from Senge’s book. The meeting was called for 2pm and John came in at 2.30 and didn’t say why. John knew exactly when the meeting was to start so he deliberately came in late. John always comes in late. We can’t count on John. He’s unreliable.

Back to my first example of my presentation. What could I do instead of concluding she thinks I’m incompetent? I could ask for data. “What is your reaction to the presentation?’ Or I could test my assumption. ‘Are you bored?’ Or I can just check the observable data. ‘You’ve been quiet’. To which the answer could just as easily be ‘Yes I am taking notes, I just love your stuff’. Maybe she doesn’t think I’m incompetent after all.

For those of us who are ‘quick in thinking’ it can be a somewhat frustrating technique. I had a very smart boss many years ago who indeed was capable of jumping to the top of his Ladder of Inference extremely quickly and actually, almost always, ‘correctly’. Luckily I could follow his thinking for the most part. A number of our fellow leadership team members however, although clever people in their own right, had trouble understanding his conclusions. So as HR VP, I would coach him to talk his way down his Ladder of Inference. For him (and often for me) it was self-evident and hence he was not that happy doing it. It could be painful process. But occasionally slowing down and explaining his thinking helped us to have the mutual understanding and productive conversations we needed to have. And to make the right decisions together.

I’ve used this a lot with my teams and it’s been really useful to build mutual understanding and improve our decision making. Although I have to admit that sometimes I’ve used it rather than ask the real question in my head: How could you possibly have got to that stupid conclusion?!


  1. Mark Hartshorne

    Mark Hartshorne: Hello Gary. It’s been a while… Thanks for an interesting article and I can see how sharing the steps along this particular journey will aid understanding. I’m probably coming at this from left field and drawing dodgy conclusions as a result, but it strikes me that this also explains the subjective decision-making process adopted by interviewers in a recruitment context. And just about the only step of the 7 that can be directly influenced by the candidate is the first – the provision of experiences and data. The rest is often down to personal and subjective judgments analogous to your steps on the ladder. Recent debates on these pages regarding 1) the validity of certain ‘stock’ interview questions and 2) the hazards of generalising about social groupings just hammer home the subjectivity with which candidates have to deal.

  2. Sandra Whiles

    Sandra Whiles:Been looking as Emily Jones and I shape our work with senior teams. A useful model to open up thinking.

  3. ig

    Wow, incredible blog layout! How long have you been blogging for?
    you make blogging look easy. The overall look of your
    website is great, as well as the content!

  4. Bovenaan in Google

    After going over a number of the blog posts on your site, I seriously appreciate your technique of blogging.


Submit a Comment

Your email address will not be published. Required fields are marked *