What a mess

I have just finished reading A Perfect Mess: The Hidden Benefits of Disorder by Eric Abrahamson and David H. Freeman. I have a feeling that this is a marmite book. Some people (like me) will love it and others will hate it. I can even predict who will hate it; the people whom the book refers to as the ‘neat police’ — the people who insist on clean desk policies and colour-coded filing systems.

This book pleads the case for the potential benefits of disorder. It also highlights the hidden costs of an over-emphasis on neatness, from the expense of maintaining rigid categorisation systems to the dangers to health of obsessive cleanliness. It provides much needed support for those of us who are ‘differently-organised’ as we attempt to fend off those who seem intent on decluttering our lives.

The topics range (in a predictably messy way) from office desks to transport systems, from business to science, from education to politics.

The authors make a distinction between disorder and chaos in its scientific meaning. Chaotic systems are actually ordered, but they are unpredictable because of their level of complexity. Disorder is more about randomness.

Disorder ordered

Abrahamson and Freedman distinguish several different types of mess (and the delightful irony of someone trying to neatly categorise different types of mess is not lost on me!):

  • Clutter — elements of a system are scattered from their normally accepted positions
  • Mixture — elements are randomly ordered or combined in unexpected ways
  • Time spread — tasks or events are loosely scheduled or randomly prioritised
  • Improvisation — there is no predetermination of how things should unfold
  • Inconsistency — goals and procedures change frequently, sometimes randomly
  • Blur — categorisation schemes are loose or vague
  • Noise — processes and information are subject to unintended outside influence
  • Distraction — focus is allowed to wander between different elements
  • Bounce — levels of activity are unpredictable or not related to demands
  • Convolution — schemes are not based on straightforward logic
  • Inclusion — few filters are in place to restrict the number and type of entities in a system
  • Distortion — things are misaligned or deformed in some way that makes them non-standard

This list is likely to cause nightmares for those of a tidy disposition.

The benefits of mess

Whilst disorder may cause significant problems in some situations, the authors highlight some of the potential benefits of different types of mess.

  • Flexibility — Messy systems may adapt more quickly than more rigidly ordered systems because they require less re-ordering
  • Completeness — Messy systems may include a more diverse array of entities than more restrictive ordered systems
  • Resonance — Messy systems may be more open to influence from the environment enabling them to reach harmony more quickly
  • Invention — Messy systems allow for unpredictable combinations and the emphasis of otherwise overlooked elements
  • Efficiency — Messy systems don’t waste energy on unnecessarily maintaining order and may solve complex problems more quickly than a rigidly systematic approach (e.g. stochastic methods)
  • Robustness — Messy systems often suffer less from damage or failure (for example you can still watch a noisy analogue TV signal but a similarly noisy digital signal would be indecipherable)

The authors are not presenting a one-sided argument; they make it clear that there are as many problems with too much disorder as there are with too much order. The point they make is that many systems (especially ones involving humans) operate best with some level of disorder. This is hardly surprising, since we have evolved to survive in a world which naturally tends towards disorder. However, we like to indulge in the mistaken belief that we can control and systematise everything.

Introducing mess

While reading this book I began to think about how mess could be used to enhance the guidance/coaching process. Unsurprisingly, I came across a few examples of where it’s already being used.

Part of the sort_it on-line career development tool is a Random Job Generator (in Option Generating), which, suggests randomly chosen occupations at the press of a button. It can be really interesting to use this with a client to see how they react to the suggestions.

Another exercise I use with groups and individuals is an exercise in which I present them with a limited list of (semi-)randomly chosen occupations and ask them to imagine that these are the only options available so they have to choose three that they would consider and three that they would reject. This often leads to some very creative and revealing discussions.

Messy learning

One of the examples of a messy approach given in the book is of the Little Red Wagon pre-school. Here children are just presented with a collection of materials and allowed to play. The teachers then improvise learning based on whatever the child is doing. They don’t have a curriculum which they determine in advance — emerges as the year progresses.

I’ve often wondered why conventional wisdom on training says that you have to have clearly defined learning outcomes before you start. I often find that participants don’t know what they really want to get from a session until they start to engage with the material. Only then can you fully determine their learning needs.

I’m increasingly drawn to sessions in which I provide some material designed to stimulate the participants to discuss a topic (usually something like a case study). I then construct the session by picking up on the themes that emerge. There’s probably a name for this type of approach.


  • Can you think of other ways in which mess can help in the career development process?
  • Should our information rooms be messier?
  • Normally, I envisage career coaching as taking the mess inside the client’s head and helping to bring more structure to it. Do you think it’s a valid part of our job sometimes to mess with our clients’ heads a bit?

, , , , , ,

  1. #1 by todd I. Stark on 24 November 2010 - 14:32

    Very interesting. Thanks very much for this review.

    Do the authors say whether there is any evidence available for whether functioning in different degrees of disorder is an individual difference? That is, do some people legitimately functional better in a more disorderly environment than others? Or are the tradeoffs involved in disorder that the authors discuss intended to be general principles that apply to everyone?

    • #2 by David Winter on 25 November 2010 - 13:40

      Hi Todd

      They do discuss individual preferences for openness and structure, but the book is partly designed as a counterbalance to propaganda from the advocates of neatness, so don’t expect much in the way of evidence (just as the advocates of neatness don’t tend to offer much evidence).

      They don’t necessarily say that everyone needs a certain amount of disorder. However, they argue that many systems/organisations have an optimal level of disorder and that trying to impose too much order can have detrimental effects.

      • #3 by Todd I. Stark on 25 November 2010 - 16:45

        Thanks David. I suspect this has a lot of merit. The trick would seem to be figuring out how to best make use of organization and selective disorder for what each person needs to do and how they work best. I’m thinking the real boon would be guidelines or skills for determining that for each person and task type combination rather than sweeping generallizations about order vs. disorder.

        It reminds me of the issues around goal setting. Goals are a panacea, goals are useless, back and forth. The skills and strategies for making flexible effective use of goals as a tool when they make sense and not letting them constrain you is the key. I think the same probably goes for order.

        In complexity theory, it’s the transition space between chaos and order that gets most interesting. That makes a nice metaphor for goals and organization as well I think.

        Happy Thanksgiving, and thanks for your interesting and useful posts!!


      • #4 by David Winter on 25 November 2010 - 16:59

        Good points.

        I guess I tend to use a compensatory model. If I get the feeling that someone is limiting themselves by being a bit too rigid and over-structured, then I’m inclined to introduce a bit of randomness. If it seems that someone is being handicapped by being overly disordered I tend to introduce a bit more structure.

        Either way I do it a bit at a time and see how it works before pushing it any further.

      • #5 by Todd I. Stark on 25 November 2010 - 17:11

        I agree with you. I like compensatory and incremental strategies any time people are already doing something right (which is almost always). Starting fresh from scratch seems like a desperation strategy with a lot of risk unless we’re really stuck badly or seem to be getting nothing done at all or just need a complete reboot.

  2. #6 by Vinny on 25 November 2010 - 10:13

    Hi David,
    I have a very messy desk and things are in a vague chronological order (If I’ve seen it recently, it must be near the top of the heap). This system works for me, but when others need to find things on my desk, then it becomes very difficult because they don’t know when I last saw the bit of paper. This is why strict order can be very useful in an information library. If clients are looking for specific information, it needs to be ordered so that it is accessible.

    However, I think that having one random and disordered file within the library might also be useful for those who don’t know what they are looking for. If I wasn’t sure which sector I wanted to work in, then the categorisation in the library would work against me, and random disorder may help to spark up ideas.

    P.S. I’d love to borrow this book!

    • #7 by David Winter on 25 November 2010 - 14:05

      True. But which strict order should you use? At the moment we use the AGCAS Occupational Classification System which is a fairly arbitrary order which makes sense to librarians. But what about a structure in which careers are grouped together based on degree requirements, or competition, or use of particular skills? They would be equally (possibly more) useful to some people.

      That’s why on-line information has much more potential because you could apply several different classification schemes simultaneously.

      You can also introduce randomness. One of the things they talk about in the book is the Banana Slug search engine which adds a random word to your search string to produce interesting and unexpected results.

  3. #8 by Bill Bell on 30 November 2010 - 17:11

    David, you might be interested in the “Random Occupations” gadget I developed for iGoogle too. At each invocation it lists ten randomly chosen categories drawn from the Canadian National Occupational Classification. I’m not altogether happy with it; I’d be interested in your comments.


    Incidentally there are new species emerging near the surface of my desk.

    • #9 by David Winter on 1 December 2010 - 11:42

      Hi Bill

      I like the idea.

      Who is it aimed at? A lot of the suggested occupations seemed to be manual work.

      The suggestions seemed a bit repetitive but perhaps that’s to do with the repetitive nature of the underlying occupational data. It might be worth using a different (more selective) list of occupations.

      The different Search and Results tabs is a bit cumbersome if I want to repeatedly generate random stuff.

      Keep up the good work.


      • #10 by Bill Bell on 1 December 2010 - 14:56

        Thanks, David.

        It’s not intended for any particular population, which might make for weaknesses. It was patterned after other gadgets I wrote which also have the system of tabs. I agree with your comments and I will take them into account for the next version.

        Thanks again.


  4. #11 by Laura on 10 January 2011 - 12:46

    Thanks David
    Perfect for me currently catching up whatever takes my fancy and making scruffy notes of random ideas and statistics etc. in a seemingly very disordered way. Whilst telling myself that really I should be using quieter times for filing, ‘finishing’ and updating.

    No I shouldn’t hurrah! I shall just change how I describe my preferred activities using terms such as improvising, resonance and bouncing,

    Hadn’t come across the sort_it online career development tool pages before, would it be ok if we link to this resource?
    Uni of Bath

    • #12 by David Winter on 11 January 2011 - 18:36

      Glad to have given you an excuse a justification for your activities.

      By all means, link to sort_it.

Leave a Reply

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: