Behavior change models – suggestions

Posted by on Aug 24, 2009 in behaviour change, Copying | 11 Comments


Just about to head off to spend an afternoon with folk particularly interested in behaviour change.

Thought I'd share these few brief thoughts about the existing models

1. most start from the assumption that the individual is the right level of granulation for studying behavior (and thus behaviour change). Fine, if we were a solitary species of independent agents but (as we argue here regularly) this doesn't appear to be a good characterisation of Homo sapiens. We are a social species – more so that most of our relatives – and we do what we do in the company and under the influence of others (real or imagined). Most of human life is – as Oscar put it – a quotation from the lives of others.

2. most of the fancy models touted aren't behaviour change models at all but rather "how to change people's behaviour" models: in other words they presume that change is something generated largely by external ("exogenous") forces and (hate the word) "levers".

3. as a result most ignore the changes in behaviour that arise without external interventions (such as marketing), assuming that this cannot amount to much. Yet these changes are happening all the time in all aspects of our lives.

4. Few admit the enormous failure rate of attempts to change people's behaviour – in marketing, in public policy, in (change) management and in our daily lives. It's really hard to set out to change behaviour – far better to help the behaviour change itself, don't you think?


  1. Ian Wallace
    August 24, 2009

    Hi Mark
    Thanks for sharing your thoughts – here are some of my experiences with behaviour change.
    1. Each individual encompasses more than one identity; it’s like every individual is a micro herd. The emergence of particular identities and behaviours depends on the circumstances that the individual is in. If the behaviour of one of the identities changes, then most of the others will copy that change.
    2. Most change models do effect some change but that change is usually for the worst. Why do we always assume that change will always be good and positive? As you say change models are always focused on the individual. Instead of focusing on the individual, it is usually far more effective to explore the context of the situation around the individual. This context will reflect the unrealised and unspoken intentions of the individual’s family of identities.
    3. The most profound changes in individual behaviour always seem to emerge indirectly from chance and serendipity rather than by direct influence and command. The more agency that the individual feels they that have in effecting a change, the more powerful and committed that change will be.
    4. Most behavioural change programmes have the reverse effect to that which was intended. The unspoken identities that we all have tend to behave like unruly teenagers. When someone tries to change their behaviour, they will automatically rebel, even though the proposed change might be quite beneficial for them.
    In my experience, behaviour change programmes rarely work. What I invariably find to be interesting and valuable is why the instigators of change are trying to change individual behaviour. Often the behaviour change that the instigators are trying to cause is one that they are resisting themselves in some way.
    If the instigators don’t change their behaviour then the people they are trying to influence have nothing to copy. However, if they can articulate intentions, values and beliefs that are currently dormant and unrealised among an individual’s family of identities, then there is a very good chance that the individual will copy that behaviour.
    This will not seem like copying to the individual or that their behaviour has been influenced in any way; it just seems like they have become more aware of who they are and what they need.
    Can you share why the folks are interested in behaviour change? If it’s all a bit confidential and hush-hush that’s fine.

  2. Kirk Fisher
    August 25, 2009

    Many change management models ignore the idea that to change we have to dissolve. We want to believe that things will remain the same while we radically change, and quickly. It’s a surgical model.

  3. miro
    September 2, 2009

    To rely on one dimension for behavior change is a recipe for disappointment.
    humans are guided by individual decisions as much as herd influences, but we have volition.
    we choose to follow, to do what is habitual, to do what is convenient, to do what is individual … the role of the herd is to establish the behavior as an acceptable social norm.
    Herds develop for different reasons, sometimes it’s in support of a preferred/optimal solution (MADD) other times there is no inherent ‘better than’ preference rather there is no tangible lasting cost/benefit of making a different decision (tastes great/less filling) and so momentum prevails offset temporarily by marketing inducements.
    If we were herd automatons there would be rapid consolidation/homogenization since any slight perturbation would result in a cascade since percolation constraints (the technical constraints to achieving a global phase change) wouldn’t apply over a large enough time span.
    To some extent we see that reflected in real life judging by the Pareto/Longtails around us
    yet at the same time we see innovation and disruption in the quantum foam
    So I suppose the only answer to the question is yes/no/maybe/sometimes
    PS Your readers might find value in “the measurement of whispers” posted on my blog – comments welcome

    September 2, 2009

    You are on the right track on all four counts as far as I am concerned!
    Some readers might be interested in the idea of behavioral design and similar concepts I talked about here –
    Yes, old wine for a new generation.

  5. Alex, aka SocialButterfly
    September 2, 2009

    I’m curious. After reading about the failures of some models, what behavior change models do you think hit the mark the closest? Are there certain behavior change models that you prefer?

  6. MisEntropy
    September 5, 2009

    Plannersphere Top 20/20 – September 2009

    The top 20 planning blogs in September 2009 and the top 20 planning blog posts from August 2009.

  7. Mark Earls
    September 8, 2009

    Thanks, Miro. I think you seem to overestimate the role of conscious or directed following. An alternative reading of the literature would suggest that much of the time we do what those around us are doing, think what they’re thinking and feel what they’re feeling. Our lives, as Wilde put it, are quotations from the lives of others – even if it seems otherwise to each of us.

  8. MariaInes
    September 23, 2009

    I am very interested in the topic, but I am not an expert at all. Some questions have came to mind in any case:
    It is easy to accept that we are herd behaved animals. You can really see it everyday in yourself and among the others.
    My questions are: Why a person with ‘radical’ values do not offen follow herds?
    Is this kind of behavior the one that allows to create ‘races’, cultures and nations?
    Are then racism/nationalism natural? and should we accept them? or maybe herd behaving is as well the seed that can replace those manifestations seen like negative from our less animal nature point of view?
    Very interesting indeed!

  9. Sue Tupling
    September 30, 2009

    It may be hard to set out to change behaviour but using the principles of Viral Change it can be done because this semi-engineered process uses the fact that we are ‘herd’ animals (See Mark Earls new book ‘Herd’) and like to follow others (as long as they have ‘influence’)
    Here’s an outline of Viral Change:
    1. A small set of behaviours has the ‘non‐linear’ power to create high impact
    2. A relatively small number of individuals within any organisation have great power in the creation of change. This power is related to various factors such as high connectivity with others, high trust, or moral, non‐hierarchical authority
    3. Behaviours endorsed and spread by that small group of individuals within an organisation (‘champions’) create ‘social tipping points’ where those new behaviours become established as a norm. ‘Critical masses’ of individuals adopting those new behaviours are created via imitation and social copying in similar ways as trends or fashions are created in the macro‐social arena
    4. The role of the formal hierarchy, management and leadership, is to support those small groups of highly influential employees who ‘infect’ the organisation with the changes they have endorsed. It is
    therefore a role largely back‐staged or ‘invisible’
    5. Peer to peer viral change works better when is spread informally (but orchestrated; we call it a process of ‘designed informality’).
    6. Viral change is neither top‐down nor strictly speaking bottom‐up, but multi‐centric and distributed across the organisation.
    7. Stories are the best currency of change. Story‐capturing and story‐telling is key to viral change.
    I’m sure you have come across Viral Change. It is a powerful amalgamation of social network theory, the law of connectivity and influence and tipping point phenomena.

  10. Geoff Brown
    October 1, 2009

    Thanks Mark
    A few years ago I ran a behaviour change pilot project called Castlemaine 500. Sounds like a car race … maybe? The goal was to change the behaviour of 500 households (in a town of 3000 households) which would result in energy reduction by 15 to 30%.
    A long story short … the overall impact of wonderfully facilitated workshops and professionally orchestrated home energy assessments was a big fat ZERO … no measurable impact when we compared our group to the broader population!
    Why? As you pointed out … CHANGE is complex, unpredictable and happens in a broader social context. It can also be lonnggg term beyond the reach of the project budget.
    If we had our time again we might do things like …
    *do much more to understand and make use of existing social networks
    *probe a bit and watch what trends are taking off already before we introduce our ‘thing’
    *put smart meters in households so we can ‘notice’ any changes in energy use as they happen … and not have to wait 12 months to compare seasonal data on energy bills
    *do much more to build the local ownership and coordination of the project very early one
    By the way, our report did admit the big failure AND also highlighted many of the positive community building and leadership outcomes that came from participants.
    Your work was used in the recommendations section … along with other Herd Thinkers like Dave Snowden.

  11. Kleptn
    April 14, 2010

    Re your 2nd point – my impression and personal experience is that a behavioural change needs to be based on some form of ‘awakening’.
    Take the example of quitting smoking – a lot of people are trying and stating that they want to change the behaviour, but are always falling back to old habits – hence no change. Those who succeed, need to have a deeply felt urge for change – based on some sort of awakening (i.e that smoking serves no purpose and you’re better off without it – REALLY better off without it.)
    My later personal experience comes from management style, where I previously considered myself a controller, wanting to make sure that everybody in my project delivered, and controlling on a very detailed level. Introducing new project methodolgy based on a very large extent of trust to the team members, our experience was that the team delivered far better that having me hanging over their shoulders. This was my ‘awakening’ and I started getting interested in pursuing what is the driving force for people, instead of focusing on mere deliverables.
    And – yes. I am a former smoker which havent touched a cigarette in more or less 10 years, and have had no urge whatsoever of starting again…
    I think also this can relate to other behavioural changes, such as teenagers ‘growing up’, or poeple suddenly becoming religious, or switching political beliefs.