Questions as Boundaries

How do we know when we are asking a good or a poor question? One sign of a poor question is when people are spending a lot of effort for seemingly little gain.

Question‘What should we do?’ can be both.

If it’s been done before and there is a predictable, best practice way to do it, then the obvious answer is ‘The next logical step.’

Obvious StepsHowever, if we are exploring complexity it can be a poor question, there are so many possible options that it can take a long time to decide what to do.

Complex SpaceIn order to explore something which has limited certainty, it is better to form a view or hypothesis and test it. This then creates a type of boundary around the thing we are trying to understand to help us make sense of it.

Questions as BoundariesQuestions like

  • Will ‘X’ happen if I do ‘Y’?
  • Is it feasible to achieve ‘X’ for ‘Y’ effort?
  • Will people buy our product?

For example, we often start a project by asking how much time and money it will take to make our idea a reality. This can drive a very large amount of effort to find out – when the project happens to be in the complex domain (as per the Cynefin Framework). If we change the question to ‘Is it feasible to make this idea a reality for $X and Y time-frame?’ it puts boundaries around the initial exploration stage and we can avoid large amounts of wasted and unfocussed effort. An example can be found in my previous post ‘Why do we Estimate?’

Questions on Cynefin FrameworkIn summary, check which type of system the question relates to according to the Cynefin Framework. Then check if the question is helping to drive useful logical work or useful exploration and if not, then change the question.

LeanUX 14

I’ve spent the last week writing up my notes from the excellent LeanUX NYC Conference held on the 10th-12th April 2014. The great news is that all of the speakers were recorded and these will be progressively released on the LeanUX NYC site. Below are just a few things that I enjoyed learning about.

Interactivism – a  model of planning where we ask what we would want to design right now that would work – instead of an ideal future state. We then identify the constraints impeding us from achieving that now and manage those constraints. From the opening keynote by Jabe Bloom.

Alistair Croll mentioned that innovation is the opposite of what the large company is designed to do and therefore innovators are seen as things like bad listeners and job killers etc…

The three things that lead to success are

  1. Intention – manifested as a decision
  2. Process – a means/routine
  3. Practice – right practice, deliberate practice

This is from John Shook who spoke about Lean Change

Thomas Wendt mentioned that testing can reveal coping strategies – things that people do to cope with poor design – I really liked this insight.

In a complex adaptive system, messy resilience is better than neat stability. The idea of boundary conditions as membranes not walls. These two items from Alicia Juarrero.

Bill Beard talked about branding moments such as making a wrong password failure a bit more fun instead of the plain ‘invalid password’ response – but only to do these sorts of things if your product is not broken. A very important point.

A complex adaptive system is modulated – not driven. It is non-causal and is dispositional. Also, frameworks are scaffolding and are meant to be taken away. These are from Dave Snowden who provided an excellent closing to the first day.

Markus Andrezak pointed out that the constraints applied to a work environment meant for creation and creativity would be harmful to a production work environment and vice versa.

The above are a sampler of the take-aways that I am finding useful to ponder about – and I would encourage you to watch the videos when they are posted. The conference was intense and informative and a great way to meet people who are leading our thinking about improving the ways that we work.





Thank you to Torbjörn Gyllebring and Thom Leggett for the conversations and inspiration to write this post.

It can be useful to step back from our work and study the decisions that we are making. Here are some of the ways to study decisions and how we can start to visualise the information.

Mapping Decisions

Start by writing out the decisions and laying them out in a logical sequence – grouping similar ones together and consequential ones after each other. Draw some arrows between them – it can be helpful if another person is working with you to help clarify which decisions lead to others and their inter-relationships.

Sometimes there are loops such as the ones labelled ‘A’ and ‘B’ where, when we make decsion A, it changes how we should make decsion B which then impacts how we should make decision A. It might be possible that we get false ‘loops’ as well as the real ones when we are not clear about the decisions that we are making.

 Decisions pull info

Once a decision has been identified, we start seeking the information required in order to make the decision. It is as if the decision is ‘pulling’ the information into it, and once there is enough, we can then make the decision. We can visualise this by writing the decision as a question on the top of a card and the information items required as bullet points under the question.

 Decision Cards

Some information items are actually prior decisions, we can use a code to link these cards together and put them up on a wall to show the realtionships. This allows us to have good conversations about both the activities required to gather information and the progress of the decisions that we are making.


New Skills

I have been knitting a pair of socks for myself this last week – on a set of four needles – in the round. It is really easy to do because I can just keep knitting and only need to use purl stitches when I make the heel.

I first tried making socks on a pair of needles (like normal knitting).

2 needlesI was very surprised when it was quite difficult – heels had to be formed at both ends, every second row was purl and the finished sock had a seam underneath the foot.

So I tried with 4 needles and it took a little while to work out how to hold them – but it is much easier.

Knitting in the roundI wonder how many other things I have avoided trying because they look difficult? I could be missing out on gaining some new skills.


It is a beautiful Sunday morning and I was reflecting about the way I used to perceive time.

I used to see the weekends as jewels that were very far apart on a string of grey and dull time that included the obligations of work in between. I would despair that the weekends were so short and start to feel stressed from Sunday lunchtime about how little I had achieved in the weekend.

That was a long time ago – now I see things very differently.

Each day is a gift and I am curious to find out what I will learn from every single day. The reason this has changed is because of the joy that I have in learning new things and having the freedom to do this at work as well as privately.

I am grateful to all the people that have helped me to understand new ways of working related to Cynefin, agile, lean and beyond. You know who you are – if we have ever spoken/tweeted around these topics, or if you have written posts, tweets or books that have inspired the people I have spoken to…Thank you.


Complicated Kanban?

Catherine Walker has asked if ‘…professional designation edges the Kanban community towards the Complicated domain…’ in a recent post. On a slightly different tangent, I think that aspects of Kanban sit in each of the Simple, Complicated and Complex Cynefin domains and the complicated aspects are the ones that could be used for certification purposes if needed.

Here is an explanation using a Kanban board as a simplistic example – of course, Kanban is much more than a board…..

Simple BoardSimple Domain – Simple Board

Setting up a board to visualise the work is easy – anyone can put up 3 columns and place work items into ‘to do’, ‘doing’ or ‘done’. This puts it into the simple domain.

Psychology of the BoardPsychology of the Board

Understanding the reasons why visualising the work on a Kanban board helps us in  many human ways takes education and expertise. Thus it belongs mostly in the complicated domain. This is an example of an area that could focus on certification – where patterns have stabilised and outcomes are predictable.

Speaking BoardBoard Informing Work

One of the most powerful things about the simple Kanban board is how it can show us ways that the work is flowing (or not) so that we can improve the ways we work. What the board will show us will be different for each context that it is used in and thus sits in the complex domain.





Thanks to an ‘off the cuff’ comment from @semanticwill about ladders of inference, I started to think about it a bit more deeply.

We each have a rich set of different experiences in our lives that make us who we are. When we make assumptions about statements that others make, we can ‘race’ up our ladders of inference – this can happen on both sides of the conversation.

Our Ladders of Inference are Uniquely Our Own

We are advised to ‘walk back down’ our ladders of inference when we make assumptions, but what I had not thought about much was that some ladders can be short and others long. So if I am working with someone else rung by rung to unpack our assumptions, I should not be surprised if one of us only goes a couple of steps and the other many more.

Causal Chains

The method for Multi-Hypothesis research that Jabe Bloom describes in his Failing Well session is very useful for exploring ideas and gaining new insights to problems.

The main idea is to use ambiguity by presenting factual statements to a group and allowing each person to form their own opinions and conclusions about those facts.

Causal Chains

We ‘unpack’ what thoughts may have led to the original opinions and conclusions, some thoughts will be certainties that the facts are right or wrong and others will be guesses and doubts.

  • Guesses and doubts are then explored to find ways that we can conduct tests or experiments in order to learn – the focus being on the smallest effort we can invest in order to learn something useful, regardless of the test failing or succeeding.
  • Certainties are sometimes worth testing as well – in the picture above, we try to invalidate gravity by throwing a ball – if it did not fall, we would be surprised and have a great opportunity for learning.

This workshop method can be completed in as little as 60 minutes with a small group, 90 minutes is comfortable for a group of about 10 people. It is a great way to get a lot of ideas in a short time and to shed some biases in our thinking by allowing many different points of view.


There are many ways to improve the way we work.


Agile, Lean, Cynefin, Lean Startup, Kanban, XP, TDD, BDD, Srcum – these are just a few.

There are also many ways to apply these approaches to the way we work.

Ways to Approach

As a ‘pure’ method, a set of principles, a staring point, assembling a mixture of approaches appropriate to the problem we are trying to solve.

The great thing is that it gives us many combinations to try so that we can find solutions that suit the outcomes we are aiming for.

The down side is that it can be very hard to choose an approach – what has worked for one place, can be difficult to apply to another and get the same outcomes.

Learning in the Cynefin Domains

If you are not familiar with the Cynefin Framework by Dave Snowden, you might want to have a quick overview by watching The Cynefin Framework

Different management methods work better in some domains compared with others, so I was wondering if different learning styles might also be more effective in some domains than others.

Learning mapped onto the Cynefin Domains

Notice that the lower two domains, Chaos and Simple, focus on learning by repetition and the upper two domains, Complex and Complicated, focus on higher forms of learning (and unlearning).

In disorder, meta-learning might be more useful or perhaps the act of learning is one of the things that pull us out of Disorder into the other domains.

Learning in the Complicated domain is how we develop our expertise – the more patterns and interactions we can learn about, the better we get at predicting how to get good outcomes and avoiding re-work and waste.

Unlearning is more important in the Complex domain – the more patterns and predictions we are good at, the less likely we are to notice the emergence of something new.