So you are sitting with your client and dropping into the conversation you are listening to you hear the phrase ‘I think’ or ‘we think’ or even ‘they think’. You are hearing a client framing an uncertain assumption. Are they possibly building an entire set of positions and feelings on a very subjective and restricted interpretation. Maybe it is time to challenge, to find the tipping point in the session – how does what you are hearing fit with the other data you have been accumulating? Is it time to test what is meant by the use in this context of that word ‘think’? And how do you help turn ‘think’ into ‘know’? And what exactly do we mean by ‘know’ – how much precision can we attribute to that confident statement? On a scale of 1 to 10 where 1 is complete uncertainty and 10 is complete empirically robust certainty where does the client place themselves?

For those of you who navigate at sea (before the advent of GPS) one of the more accurate ways of finding your position when within sight of shore was to take a series of bearings on known features, turn them into back-bearings and then plot the result on the map. Normally a minimum of three points are required for a good fix. However inevitably, given the motion of the ship, weather conditions, the compass and other variables the result was not a neat intersection of three lines at a single point.  Instead there was a triangular area of sorts created by the three lines. You ‘knew’ that the ship was somewhere within that triangle – or otherwise known as a cocked hat because of its resemblance to the naval headgear worn at the time. Hence the term ‘knocked into a cocked hat’.

So how do we help the client take the bearings that help him change the balance from ‘think’ to ‘know’? The most common approach for work based behavioural and performance issues is the 360 degree questionnaire. However for other issues, perhaps the performance of a project or the views of a colleague, this is unlikely to be appropriate or possible. However there is a simple 360 degree exercise you can run during a session.

So a suggestion. Explain the cocked hat model to the client (at least they might learn something fun from the session for their next pub quiz). Get the client to frame the ‘I think’ statement as a broader open second order hypothesis. Instead of ‘I think this project is struggling’ turn it into a second order question such as ‘how do I understand better how this project is performing?’ Avoid the temptation to ask the first order question that simply reinforces the initial assumption ‘how do I find out why this project is performing poorly?’

Get the client to identify the points from which he/she needs to plot the position. These may be other people (peers, internal workforce, and clients), performance data and also themselves. Try and push for a good spread – good positions are derived when the reference points are well spread. Place a token on the floor or table which represents the hypothesis. Place further tokens that represent the other points that have been identified around the hypothesis. Get the client to position them in relation to the hypothesis token and each other. Maybe you have the time for the client to draw a quick picture of on each token. Is there any significance in the placing? Does a series of closely placed tokens mean that you need to find one with a more distinctive position away from the others? Where does the client’s own token sit in relation to the other tokens?

Then get the client to physically move to each token. Get them to sit quietly for a moment and visualise and then describe the individual they are representing. Ask them to describe how the individual might see and answer the hypothesis; how they are related to it, what they may feel about it and what ‘data’ (qualitative and quantitative) they might have to contribute. Use a brief but intensive ‘T’ model of questioning at each point.

After each position ask the client what they might need to ask of the individual whose token they are at to help them better understand that particular perspective on the issue. Get them to jot down the key points on a sheet or post-it next to each token before they move on to the next position.

Finally get the client to move to hypothesis token. Get them to summarise the position and questions to be asked for each of the other tokens. Make sure they face each token as they do so.

On a scale of 1 to 10 how more certain will they be about the project performance and its issues when they have talked to their reference points? What is the cost of not asking those questions? What are the benefits of asking those questions?

Of course you can use this technique to test the ‘know’ statement as well. In this case ask for the confidence rating at the start – on a scale of 1 to 10 how certain is the ‘know’. And do the same at the end. Do not be surprised if the rating is lower - indeed a lower rating can be celebrated, as it shows insight and the willingness to stand back and walk around an issue.

I would be interested in hearing from anyone who uses this technique. Let me know how it goes. Anything you could suggest to improve it?