Tag Archives: Cynefin

Metaphors are about communication, not truth

If a friend says to you:

“I just left my terrible job. I don’t know why I stayed so long. I guess I was just like a slowly boiled frog”

Do you get her meaning?[1]

Do you know that the metaphor is not true?[2]

Does it matter?[3]


  1. That the changes were so slow and gradual that she barely noticed them and thus stayed well past the point she should have done.  ↩

  2. The metaphor is that “If you put a frog in a pan of already boiling water, it will jump out. But if you place it in a cold pan and then slowly heat it, it will sit there calmly until it boils to death”

    Except it’s a complete myth that a frog won’t notice and allow itself to be boiled alive. The fact that this myth is so widespread however I take as a good thing, because that means there are very few people wanting to test it out. Although sadly this was not always true:

    Several experiments involving recording the reaction of frogs to slowly heated water took place in the 19th century. In 1869, while doing experiments searching for the location of the soul, German physiologist Friedrich Goltz demonstrated that a frog that has had its brain removed will remain in slowly heated water, but an intact frog attempted to escape the water when it reached 25 °C.

    The moral of which I’m guessing is you can slowly boil a frog anything as long as you remove its brain.

    Additionally the entire metaphor is silly if you think about it – because a frog dumped in boiling water would not jump out – it would die.  ↩

  3. IMHO? No. Because we’ve mangaged to convey a concept clearly and concisely – typically avoiding all the nonsense contained in fn2 above.

    Bottom line here is that if people are using a metaphor to describe something to you – focus on the communication aspect rather than picking apart the factual accuracy of the thing.  ↩

Best for me, best for now (Practice)

In my last post, I covered the concept of “Best for Me” Practice.

In this post I want to extend that a little further into “Best for Me, Best for Now” (Practice).

The Tyranny of “Best”

“Best Practice” is often held up as a gold standard – something to aspire to, rather than to do.[1]

This leads broadly to two dysfunctions:

  1. Aiming too high
  2. Aiming too low

Aiming too high is akin to somebody who desires to increase their fitness through running and reads that “best practice” for fitness running is to run 5km, in under 45 minutes three or more times a week. They walk out onto the street, never having run a step in their life and hop to it.

Aiming too low is akin to somebody who reads the same advice and decides that “well, given that I’m 80kg overweight and struggle to walk down to the shops, I could never do that.”

Neither case is likely to ever reap the benefit. (And if it’s not obvious, aiming too high can rapidly lead to aiming too low, with a short bout of depression and self loathing in the middle)

Focussing on Outcomes, not Activities

Ultimately the real problem with Best Practice, is that it places the focus on the means and not the ends. There is a massive leap of faith that by doing this, we’ll get that. And if you’re not getting this fast enough (or at all) then just do that harder.

Rather than trying to run 15km a week, both our hypothetical couch potatoes would be better off taking on board a “Best for Me, Best for Now” approach.

The outcome they want is a longer healthier life; and the best first step is probably along the lines of:

  1. Walking a little more than they currently do
  2. Making some dietary changes
  3. Adding other physical activities that they enjoy and will sustain and won’t cause injury [2]
  4. Tracking some metrics – looking for correlations[3]

“Best” Practices, like MVP’s[4] should be considered as hypothesis

If you did those four things – then you would rapidly[5] become a different person with a different capability (and potentially even different goals).

And thus Best for you, Best for now would change.

And that’s a good thing.


  1. And if it is seen as something to do, then often the question turns to why aren’t you doing it now?!  ↩

  2. When you focus on the outcome, you free yourself up to explore other alternate paths to reaching it. Running is a way to get exercise, it’s not the only way. Considering that when it comes to weight loss, exercise is more about improving mood than it is about burning calories, it’s doubly pointless to do something that you don’t enjoy.  ↩

  3. Perhaps measuring this with a FitBit and some bathroom scales. Noticing that on the weeks they walk more, they lose or at the very least maintain their weight and feel better to boot.  ↩

  4. Minimal Viable Product  ↩

  5. OK, this would depend also on your age, starting weight and whether or not you had any pre-existing glandular or metabolic conditions, but hopefully you get my point.  ↩

Best (for me) Practice

What do we mean when we speak of Best Practice?

Best Practice is a rather polarising term. Very few people are neutral about it. It’s either the watchword of an organisation or an instant irritant. But what does it actually mean?

Genuine Best Practice

As previously discussed, in Authentically Simple Systems – the term “Best Practice” is entirely valid. Given the current body of knowledge, there is One Best Way to perform this task.

To do otherwise would be, by definition, either inefficient or unecessarily prone to error.

But this term is used widely outside of the Simple Domain; and not ironically.

Best (for me) Practice

A lot of the time “Best Practice” is really just a shorthand for “Best for Me” Practice.

But this shorthand can actually hide two completely different intents…

Best (for me) Practice, given that I don’t want to take responsibility for my own actions

This is, at least for me, is the darker and more frustrating meaning I’ve encountered.

It often equates to “just tell me what to do”, with the subtle undertone of:

“If this doesn’t work, then it’s not my fault, because you told me it would work. I didn’t fail, the practice did. And so did you for recommending it.”[1]

Best (for me) Practice, given my context and your knowledge

If you’re not a Complexity Geek, then you’re probably not as hung up on the term “Best Practice” as much as some people are.[2] To you it infers something actually far more wooly, subtle and frankly more complex[3]:

“Given everything you’ve seen about our situation so far, combined with everything you know and your general experience, what are the best recommendations you have for us?”

Which to me, seems fair enough. Even if I still rankle occasionally at the term.

So how can we tell the difference? (And also avoid a pointless argument about nomenclature?)

Practice or Principle?

One aspect of the concept of “Best Practice” that is I think universal (regardless of the meaning you adhere to the term), is that a Best Practice is something that’s been tried before by somebody else[4]

The logic is sound (if not overly courageous):

Somebody else has already taken this risk, figured this out and now I’m going to reap all the rewards, while taking none of the risk.

This all seems fair and theoretically puts a new spin on our two “Best for Me” cases above, effectively merging them. Both are simply risk adverse and wanting help.

But now we’re back to square one – because it’s only for systems that live in the Simple Space that cause and effect is infinitely repeatable.[5]

And if that’s true, then Best Practice may not be as simple as it seems after all[6]

Fundamental Attribution Error and confusing correlation with causality

Just because somebody has tried something before and they were successful (even repeatedly so), doesn’t mean that it was those practices in particular that caused (or even contributed) to their success.[7]

As science progresses, every day we’re discovering new evidence that much of what we considered causal is now correlated at best, and in some cases completely unrelated.[8]

Best Principles? (or how to tell the difference)

And this is how you can tell the difference between a “Best for me in my context” and a “Best for not taking any responsibility” intent.

Those folks who are genuinely interested in better outcomes for their problems in their context will take on board the concept of Best Principles – Principles and heuristics that successful organisations use in order to develop their own (evolving set of) Best Practices.

And those who are interested only in shirking responsibility? They’ll listen politely and then quietly insist that you tell them what to do.[9]


  1. You could argue that this is precisely why branded methods are so popular. They’re less about buying a solution than they are about acquiring a scapegoat.  ↩

  2. Also, well done for reading this far.  ↩

  3. Ironic no?  ↩

  4. And hopefully shown to work. because otherwise we’ll have to class “Tilting at Windmills” as Best Practice too.  ↩

  5. To make this clearer at the very least you’d have to be doing exactly what the other organisation was doing – in which case I would question where your competitive advantage was coming from. But even then, you’re almost certainly doing it with a different set of people.  ↩

  6. Pun partially intended.  ↩

  7. I’ve met the odd entrepreneur that attributes Apple’s success directly to the less cuddly parts of Steve Job’s personality; apparently nobody else at Apple does jack squat.  ↩

  8. By which I mean there is no scientific evidence to support any kind of relationship whatsoever; however likely or plausible it might seem to the layperson that there is a link. This seems especially true of dietary advice.  ↩

  9. Whether or not this is because of an innate character flaw or simply that this is what they’ve been incentivised to do is a topic for another day.  ↩

Best Practice is not a panacea

Best Practice is a term that’s bandied around a lot these days.

It ascribes to wishful thinking, and speaks to our fundamental human fear of the true nature of reality.

The very idea that there is “one right solution” helps provide the illusion that the world is a well ordered place.  Or at the very least could be, if we just tried hard enough.

However, as students of complexity are well aware, the only legitimate domain to apply “Best Practice” is in the Simple Domain.  This is a domain where Cause and Effect are obvious to all, and also endlessly repeatable. (And as such prime candidates for codification)

607px Cynefin framework Feb 2011

But most of life, especially knowledge work, does not fall into this appealingly controllable space, as much as we might like it to.

To increase our feeling of control, we often like to codify our notions of Best Practice into rules.  After all, if something is truly “best” then all other options are inferior and should therefore be avoided.

We then distribute these rules out to the population of humans whose behaviour we wish to control and sleep soundly knowing that our job is done and the world will be a better place from now on.

Because of course, everybody will follow the rules, and thus both their behaviour and the outcomes that they produce will therefore be “optimal”.

But what if we don’t trust people to follow our rules?

Why then, for the people’s own good (remember we we’re talking about the one best way here) we select some people that we do trust and have them enforce both our rules and their precious Best Practice payloads.

In the average corporation we might call these Best Practices by other names such as “policies”, “rules” or “procedures” and we call our trusted caste “managers” which we can then arrange in a hierarchy so that spatial position corresponds to the level of trust we place in each of them.  Everything is now ship shape and Bristol Fashion!

However, as I’m sure we’ve all experienced first hand.  The human mind likes to game the system.  We do so for all manner of reason: for fun, for malice, for personal gain and oftentimes just to get our job done.

The map does not always match the territory.  Maybe the rule is outdated, inapplicable to our current circumstance or simply badly or meanly written. Regardless, if it conflicts with our superordinate purpose then it gets in the way of the creation of value and betterment.

Regardless of the motivation, the instant a rule is created, people will be looking for ways to get around it, but without risking censure.

The unspoken mandate then becomes to follow the letter, but not the spirit of the rule, and therefore the same applies to the application of the “best” practice it was designed to engender.

This can lead to an arms race between the rule makers and the rule breakers. And so, rather than simply having one rule to follow, we have many.  Each describing the “Best Practice” version of the events that our incoveniently highly variable universe has chosen to throw at us.  Each created in retrospect for a situation which may never arise again.

Which brings us all the way back to the concept of attribution error that I spoke about here.

The rule makers’ attribution error is almost certainly that they believe that every problem is Simple and that Best Practice is therefore universally applicable to every situation.

If that’s your belief, then when you discover that the rules are not in fact engendering your desired outcomes, your reaction to fixing the situation is almost certainly to add more rules, or find ways to enforce the ones you have.  It will keep you busy, that’s for sure, it might even keep you entertained.

But it’s not going to give you better results.

If it’s the wrong tool for the job, it’s irrelevant how skilled you are at using it.

You cannot fix a broken watch with a chainsaw.

But once you see the world for what it is.  That it can be complicated, complex and chaotic, then you can achieve success through applying methods appropriate for managing those domains and give your rulebook a much needed rest.

Complexity: Justify, Explain or Apply?

Any idea, theory or science can be used to:

  1. Justify what you’re doing now, or would like to start doing
  2. Explain previous events, why some things fail and others succeeded
  3. Applied to create or synthesise new approaches to problems you have today

All are valid, and do not exist in isolation; rather they exist as overlapping zones in a continuum , and you can move between them at will.

Complexity Science’s oldest and most loyal cheerleaders are probably the Scrum Community; Complexity Science gets a mention in Ken’s landmark book “Agile Software Development with Scrum” – and the Stacey Matrix appears in the majority of Scrum courses today.  Scrum is touted as “a Framework for Product Development in the Complex Space” and the Stacy Matrix serves well to explain what the Complex Space is and why you need a new approach to dealing with it.

So this is useful, but ultimately it is simply using Complexity Science as a means to justify changing the approach used to manage Software Product Development (or anything else that Scrum is being applied to these days)

Nobody sane makes a change without some kind of Justification,  so this has been and still is a valid and useful technique.

Problems can occur however thanks to our friend Confirmation Bias.  If you like Scrum then you’ll want to use it all the time, and because of this you’ll be biased  towards seeing every problem as “Complex” so that you can use your favourite tool on it.  You have effectively created a tiny mental prison for yourself.

“So what!?”, you might ask…

in which case I would reply: Scrum has been around for 15 years or so, and we’ve been attaching the “Agile” label to things for the last 10, and the results have been rather varied.  Panacea it most certainly isn’t.

But why?

This sort of question has led myself (and others) to delve deeper into the techniques we’ve been using, the underlying theories (if any) they were based on, and to seek out the latest research in a quest to truly understand what is going on, and what expectations we should have.

This search led me through a personal journey beyond Stacey and into the wonderful (but hard to digest) world of the late Max Boisot and his I-Space model and then ultimately onto the current cool kid on the Complexity Block Dave Snowden’s Cynefin.  I then delved into works of Baumeister, Kahneman, Tversky et al to better understand the limits of our cognition, how we make decisions and effective mechanisms to overcome procrastination.

I used this research to understand what the Scrum practices were trying to achieve, and what effects they actually had. I then extended this knowledge to create a model to explain why Scrum (as interpreted solely by its practices) will thrive in certain situations and self immolate in others.   I presented these findings in a workshop I held at Agile 2011 called “Scrum as a Thinking Toolkit”

This is just one example of moving from Justification (where we know the right answer and seek information only to confirm our existing beliefs and desired actions) – towards explanation – the creation of models & theories which help us understand why things have historically gone the way they did and thus hopefully provides useful insights into dealing with the future.

Once you have begun to understand, you can then begin to Apply.  Once you know the “Why” you can begin to create your own “How” – and even better, you can  take advantage of the considerable advances that have been made in the area of human cognition and social complexity to enhance your work.  As such, among other things, I began to apply “SenseMaking” to the management of Backlogs and Ritual Dissent to Release Planning.

And you know what I found?

Hey this stuff works…

This is the philosophy that is at the root of CALM (Complexity Agile Lean Mashups) – that of “theory informed practice” – synthesise new practices based on well researched principles and then validate these practices in the field.  Loop back and continue.

It’s this level of understanding that will help you understand why in some projects removing the time boxes made things improve, whilst in others it led to anarchy and disaster.

So I can imagine what some of you are currently thinking:

“Hey! Complexity Science was originally applied to create Scrum in the first place!  It’s built in! I don’t need this!”

And that’s exactly my point.  Regardless of where we mark the Scrum epoch (somewhere between 1987 and 1994 I believe is the currently accepted range) – Science has moved on.  Many of the models used to originally describe (and thus by association define) Scrum are now dated at best and debunked at worst.  We know so much more now.  The heartening thing is that using our current best available knowledge the “old girl” actually stands up very well for her stated mission.

But there is still more to have – much of software today exists outside of the Complex realm, there are other containers than time boxes, there is knowledge out there that scales far beyond the methods we’re using today; and perhaps most importantly there are strategies and techniques which can help us address the wider organisational issues beyond just those we encounter in our teams and on our projects.

CALM is about making a move from using science to justify what we already want to do, towards a mindful application of theory; a cycle of evolution and synthesise where practice and theory co-evolve to help us realise our true potential.

The first CALM event, CALM Alpha will be held in the UK on the 16th and 17th of Feb 2012.