Cynefin, concept work, and the role of deliberate practice

Over the past week or so there have been several blogs that have helped me pull together a bunch of things I’ve been trying to connect in my mind for a while.

First was Harold Jarche’s post Working Together, in which he looked at Shawn Callahan’s ideas on group work against the backdrop of Tom Haskins discussion of the Cynefin and TIMN frameworks. Next was Tony Karrer and Ken Allan‘s discussion of the role of deliberate practice in the development of skills less than that of an expert, based on Tony’s question:

Any thoughts on how deliberative practice relates to becoming something less than an expert.  It seems it should be applicable to all levels of achievement, but everything I’m reading is the study of becoming an expert.  Is that just aspirational, or is deliberative practice also studied for quick attainment of proficiency?

Read Tony and Ken’s posts, along with the comments, for all the discussion including my comment:

…the application of deliberate practice is not the most efficient way to achieve basic proficiency, even though it would be effective. As proficiency turns into literacy and then mastery, I think that deliberate practice becomes not just the most effective way but the most efficient as well.

After some thought, and several pages of scribbles, scratches, and doodles in my notebook, I put together the following table that pulls together several different topics using Cynefin as a guide.


The first two columns come directly from the definition of the Cynefin framework. I had just a bit of trouble in the third column, primarily in trying to figure out what the best term would be to carry out “simple” work tasks.  I’m not completely happy with the term “assembly line”, but I think it gets the idea across. I am open to any suggestions to improve this.

I was also not quite sure about the use of the terms in the “Skill Level” column, specifically the order of “fluency” and “literacy”.  Again, I’m interested to hear your thoughts on this.

The heart of the table, especially as it applies to the original question that Tony asked, is the column “How to Achieve”.  Various levels of deliberate practice could have been included in each row, but in looking at each level of complexity as a stand-alone level it seems to me for the “simple” and “complicated” tasks that deliberate practice, at least as defined by Geoff Colvin in “Secrets of Greatness” and the more in-depth Talent is Overrated, is overkill. And probably an unreasonable expectation to have of people who just want to do their job and go home, which is more typical of those performing this type of work.

It is once you move into the area of complex and chaotic work that the benefits gained from deliberate practice are needed, in fact necessary.  Not only must you be able to apply what is already known in ways that have already been identified, you need to be able to learn new things and figure out how to apply them in new ways. That is the nature of mastery, and the ultimate result of deliberate practice.


15 thoughts on “Cynefin, concept work, and the role of deliberate practice

  1. Hi Brett: I’m glad I’ve discovered your blog and found you’re wrestling with combining different frameworks like I am. I noticed on an earlier post your wrote that you’re aware of how Josh Waitzkin describes his process of mastering chess and T’ai Ch’i. Here’s how I’d apply the Cynefin framework to his mastery process. This will reconfigure your chart considerably if it fits with what you’re thinking about deliberative practice and proficiency.

    There was a time when Josh was caught between two Russian chess masters with very different teaching styles, pulling him in opposite directions and messing up his game. He was at a similar juncture anytime he encountered a new T’ai Ch’i opponent would could outmaneuver Josh in ways he could not observe. It happened too fast in unexpected ways. These are CHAOTIC situations where the only option is to try stuff and see what works. In the Cynefin model, these attempts are called novel practice. It’s what, I’m thinking, comes first in a process of achieving proficiency or going beyond that to mastery. Work type: improvisation, experimentation, getting oriented, clearing up confusion, gaining familiarity, etc. Skill level: beginner

    Next, Josh would get hold of some idea of how to beat an opponent in chess or T’ai Ch’i. Through trial and error, he had eliminated different strategies and arrived at the best one. He had reduced winning to a SIMPLE situation and cultivated a “best practice” that demonstrated his proficiency. This is involves “practice makes perfect” drills like “getting better at writing essays by writing essays”. It’s already decided how to win in plan. All that needs work is execution of that plan. Work type: routine, sequenced, skill-sets, drills, etc Skill level: proficient

    Josh would then find out he could be outsmarted. The game had shifted from proficiency of skills to the inner game of focus and the ability to read opponents who were playing mind games in order to intimidate him, distract him or get him to panic. This creates a COMPLICATED situation that takes more than proficiency to succeed. This calls for the deliberative practice of psyching oneself out to find personal vulnerabilities, blind spots, hang-ups, etc. Then the attempts made by opponents to take advantage of those weak points become obvious to see coming, useful bait to use against opponents, and motivation to achieve mastery. Work type: analytic, diagnostic, debugging, fine-tuning, self-evaluative, etc. Skill level: deliberative

    Finally, Josh could handle every challenge without even thinking. He could get challenged without breaking stride regardless of the very COMPLEX maneuvers imposed by opponents, referees, and championship officials. One could say his calm focused mindset, skillfulness, timing, etc — were emergent practices. They came about from his experience of himself and the situation being as complex as it really was. Work type: responsiveness, flow state, centered, etc. Skill level: mastery

    I hope this helps you organize the ideas you’re pondering and refine your chart framework.



  2. Tom,

    Thanks for the comments, and also for using Josh Waitzkin’s stories as illustration. They are very good examples, and more importantly it provides a common ground for a discussion.

    One of the things I noticed in your post about Cynefin and TIMN frameworks was that you seemed to be starting in the “chaos” quad of Cynefin and moving counterclockwise to end up in “complex” quad as the result, or culmination of activity. Your analysis here of Josh’s learning and playing seems to bear that out.

    In my day job as a Systems Engineer, I typically see – and use – Cynefin as going clockwise from “chaos” to “simple”, from “here’s the problem holy crap how are we going to solve this” to “here’s how you build the thing.” I think this is partly why I start at “simple” and work my way around to “chaos” instead of starting at “chaos” and working around to “complex” when going the other way.

    When I look at something like chess or t’ai chi, for instance, I have a systems engineering kind of approach looking back to the beginning of the game. By the time someone today starts playing, all of the rules are well known and straightforward, and this is where all beginners start out – simple. Through training, experience, and deliberate practice the player becomes better equipped to handle complicated, complex, and ultimately chaotic situations.

    Things these masters learn may get reintroduced into the basic canon of the game for the new beginners to learn, thus changing what is considered “simple.” That’s part of why it becomes so hard for new beginners to do well in old, well-established sports. The “simple” parts of the game become vast in quantity; just think of the incredibly huge library of openings a beginning chess player must memorize, just so they can survive a game long enough to make it to the part where they can show some creativity.


  3. I’ve been struggling with this a bit and am in the process of reading a few books on these concepts.

    I’m not sure I buy the how to achieve column quite yet. What do you base that on? Why are those preferred?

    Also, it seems like deliberative practice requires breaking down to specific skills and knowledge, but that is often not the case for some complex or chaotic domains.


  4. Tony,

    Thanks for the comments, and questions. I am still working all this out in my head as well, and am working on a follow up post to flesh those ideas out. In a nutshell, I took my experience from sports and the military to establish a progression of skill development that would ultimately lead to the ability to effectively handle a situation that arises in any of the 4 Cynefin domains.

    You are right that deliberate practice does entail breaking things down to a very basic level of specific skills and knowledge, and also that performance at any of the “skill levels” I included can be enhanced through deliberate practice. The main reason I added the “work type” column was to differentiate between the various types of work done in organizations and to show what process I felt was best suited (in both effectiveness and efficiency) for someone performing that type of work.

    For example, someone content to perform “information assembly line” work is likely not willing to submit to the rigors required of deliberate practice, and the organization has little incentive to provide training beyond that required for the person to do their job.

    There is no doubt that people who don’t have the benefit of deliberate practice in their chosen field often find themselves in a complex or chaotic situation. In my experience, these people are far less likely to navigate this situation as successfully (in terms of time, cost, or quality) as someone with the benefit of practice.


  5. I wonder if “procedural” works for you as an alternative to “assembly line”?

    It’s essential to follow standard procedures on an assembly line or tuned up- and down-stream processes may be adversely affected. Workers at one station may have little or no knowledge of what effect performing a non-standard procedure would have.

    These procedures would include standard handling for exception cases, like failure of a device or local supply chain.

    This would then encompass workflow in all sorts of business.


  6. Roy,

    I’ve been going through old blog articles and realized that I had not responded to your question here. My apologies for the delay.

    I think that “procedural” captures my intent very well. Having just finished Seth Godin’s latest book, Linchpin, I also think that his characterization of these types of jobs as “factory work” is a good one. He makes a very strong argument for the point that “factory work” doesn’t just happen in manufacturing, but anywhere the job is based on strictly following a set procedure.


  7. …the application of deliberate practice is not the most efficient way to achieve basic proficiency, even though it would be effective. As proficiency turns into literacy and then mastery, I think that deliberate practice becomes not just the most effective way but the most efficient as well.

    I think that this has been shown to be true in medicine. In this article:

    surgical residents trained on a virtual reality simulator (deliberative practice) were shown to make fewer errors and to operate faster than those residents trained in the traditional “on the job” training in the operating room.



  8. The work type column is important but in reality work type is not a discreet value.

    I would prefer a model that uses a “thinking index”. I call it “Cranial Index” in my work. Some work requires little thinking at execution time. Other types of work require lots of thinking.

    A scale approach will cover “assembly line or procedural”.

    I used the words “execution time” because even “assembly line” work did require thinking at one point. To come back to the chess analogy, chess openings are very procedural and require no thinking at execution time. Chess openings, at design time, requires the highest level of thinking. The concept of execution time and design time are important because a task that is “Knowledge work” can after a while actually become “procedural”.

    Mastery may very well be the point where things who were once complex become procedural!


  9. Gilbert,

    Your distinction between “execution” time and “design” time is important. Thinking back to when I first put this together, I realize that in defining the work type I was implicitly calling out the work at execution time.

    If I were to add an extra column to my table above to show your “Cranial Index”, I think that it would go from low to high as we progress from simple to complex.

    Your point raises an interesting question in my mind: If execution of concept work requires mastery, what would the design of concept work require? Or, even more bluntly, can you “design” concept work?

    Thanks for the new insight into how to look at this.


  10. Can concept work be designed? Yes.

    We rarely try to design or procedurize concept work but my thoughts on this is that it can be done. It can be quite expensive to do so because you need a lot of expertise to lay out a process.

    For example, one can structure problem solving processes in most fields. It is only effective if the executioner masters the unit skills required to do each step. The odds of finding someone mastering unit problem-solving skills is very low, so it probably wouldn’t be worth the effort of designing concept work.

    To come back to a chess analogy. A few years back, I read “Think Like a Grandmaster” and did the exercises to get into the disciplined analysis approach (a procedure). Worked wonders. I won my next chess tournament, winning every game against higher rated players, to everyone’s amazement. It was like being on mental steroids.

    A student of mine actually pointed out that what differentiates a professional from the every day person is that one uses a “Methodology”. After 30 years I would say that the Methodology differentiates the competent individual from the incomptent ones.

    I worked as a functional analyst for a while using a methodology call productivity plus. Those who mastered the unit skills and followed the step by step approach usually produced better “concepts”.

    Take project management. It is relatively high on the conceptual scale. It can be improved with procedures but the project manager must have a very deep understanding of things like managing risks, or assumptions for the procedures to be effective.

    White collar workers and ivory tower dwellers don’t usually like having the work designed for them.

    In my mind your table is leading you in a difficult direction. Probably would be easier if you went back to the basic philosophy/ideas of cynefin and worked your way up. My guess is that you would just end up with one type of graph showing obviousness of solution on horizontal axis and the vertical axis would be interchangeable things like amount of management required, type of training, etc. “If my grandmother can’t understand it easily, it probably aint gone a work”.



  11. Gilbert,

    You are right on with “It can be improved with procedures but the project manager must have a very deep understanding of things like managing risks, or assumptions for the procedures to be effective.” I’ve recently read, and written about, the book The Checklist Manifesto. I’d love to hear your thoughts on that as it seems to fit very well with the discussion we are having here.


  12. I had to smile when I saw the checklist for checklists.

    We seem to be pretty much aligned in our thoughts but here is my thought on knowledge work in general.

    Many years ago, the industrial sector didn’t understand work. It took a few geniuses like Taylor,Gilbreth,and others to revolutionize the science of work. Note that their philosophies were misused by capitalist, but the idea of studying the types of work made sense. We need these types of geniuses to step up soon.

    As the knowledge economy becomes predominant someone will have to redefine productivity and work science for these fields. At the beginning of the “Knowledge Shift” knowledge workers feel very little pressure but with time “knowledge workers” will have to become more productive. And with more knowledge workers also comes a need for productivity.

    Right now that is not an issue. You don’t have to be productivie in a high growth sector but when grow slows down productivity becomes an issue. We have already seen a lot of north american computer programming contracts go overseas. At one point knowledge workers will have to be productive.

    Networking, PKM, Web2.0, etc won’t cut it when the big changes start rolling in. We are just starting the climb towards a predominantly “knowledge based economy” and a large percentage of the population is starting to lag in skills.

    Not sure what this has to do with deliberate practice but thought I’d share my perspective on white collar productivity.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s