If you work in health promotion or sustainability, you’ll have heard of “Social Marketing” and “Community-based Social Marketing”.

Lately I’ve noticed how these communication methodologies are being treated with almost magical reverence, as if they are the long-awaited silver bullets for the complex social, health and environmental problems we all struggle with.

I believe many of the expectations being placed on Social Marketing and Community-based Social Marketing are seriously overblown and it’s time social change practitioners reassessed their attitude to these practices.

Here’s why:

Of course you can market brands. But behaviour change is not like buying a different brand of beer, it’s about getting people to DO THINGS THEY ARE UNCOMFORTABLE WITH, DON’T WANT TO DO OR CAN’T DO, or they would already be doing them. Like parents letting their kids walk to school, or smokers quitting, or drivers switching to public transport.

These kinds of social, health and environmental behaviours are intractable because they are part of complex, “wicked” or messy social problems. That’s why they are still with us. They are intractable for very good reasons: they are fixed firmly in place by a powerful matrix of institutional, technological and social factors. To be effective change programs must therefore do more than just communicate persuasive messages, they must aim to modify those factors.

Paul Stern of the Division of Behavioral and Social Sciences and Education of the UK’s National Research Council explains that many behaviours are simply not amenable to voluntary change: [i]

“This pattern of [contextual] influences implies that effective laws and regulations, strong financial incentives or penalties, irresistible technology, powerful social norms, and the like can leave little room for personal factors to affect behavior…”

In other words, when people have very little choice how they act, structural changes (like regulation, pricing, infrastructure, service provision, governance reform, social innovation, and technological innovation) should be the preferred approaches.

He goes on to say that: “[however] when contextual influences are weak, personal factors…are likely to be the strongest influence on behavior.” However, if we are realistic, there are very few situations where contextual factors are weak. Every personal decision is thoroughly embedded in its context. Even a simple voluntary behaviour like “turning off the lights” is determined by technology and pricing.

The fact is, every effective social change effort has been predominantly structural. Improving the anti-social behaviour of drinkers, for instance, has required collaboration between police, community leaders and licensing authorities; physical re-design of venues; modified management practices; training for staff; advocacy; political leadership; and legislative change. Marketing has been the least important factor in the mix. Most solutions to “wicked” problems are like this. They involve multi-faceted strategies, and are very much about building relationships and re-designing practices, places and institutions, with marketing taking an important supportive role.

Of course there’s nothing wrong with good marketing. It’s a vital part of the mix. It spreads knowledge, creates interest, helps get people buzzing, and helps spark political action so that politicians get busy with the work of changing institutions and supporting technological innovation. It is an important handmaiden of change, but not the driver.

Let’s be clear what Social Marketing is

Social Marketing [ii] (SM) is a way of planning communication programs that aim to influence human behaviour. Community-based Social Marketing [iii] (CBSM) is a variant that includes influence techniques drawn from social psychology. Communication for Behavioural Impact [iv] (COMBI) is another variant that’s been designed for developing countries.

One of the most commonly heard definitions of SM is:

“Social Marketing is the application of commercial marketing technologies to the analysis, planning, execution and evaluation of programs designed to influence the voluntary behaviour of target audiences to improve their personal welfare and that of society.”[v]

The practice of SM (and SBSM and COMBI are very similar) is said to consist of:

1) Start with a specific behavioural goal.

2) Conduct research with the target audience(s).

3) Be informed by psychological theories or models.

4) Tailor your efforts to suit the needs of the target audience(s).

5) Consider the 4 Ps: Product, Price, Place and Promotion.

6) Offer personal outcomes that the audience values.

7) Address the influence of competing promotions.

This reads like applied common sense. You wouldn’t want to design a communication campaign any other way. Perhaps the excitement that surrounds SM is partly due to what it replaced, which was a complete lack of method in the design of health promotion efforts.

What’s the evidence for Social Marketing

As far as I can find, there has been only one systematic review of Social Marketing practice.[vi] This 2007 review, funded by the UK’s National Social Marketing Centre, analysed the results of 54 Social Marketing programs focusing on alcohol, tobacco, illicit drug use and physical activity.

The researchers concluded, in part: “A majority of the [youth] interventions…reported significant positive effects in the short term. Effects tended to dissipate in the medium and longer term… These results are broadly comparable wit
h systematic reviews of other types of substance use prevention interventions
. The evidence is more mixed for adult smoking cessation, although small numbers of programs were nonetheless effective in this area.”[vii] [my emphasis]

In other words, these SM programs were found to be about as effective as interventions not based on SM methodology, which, I assume, means other kinds of educational interventions.

Notably, 48 of these 54 programs relied substantially on face-to-face tactics, like counselling and peer support, in addition to mass media. The results therefore can’t be extrapolated to the great majority of Social Marketing campaigns, which consist primarily of mass media efforts. Since face-to-face interaction generally has far greater personal impact than mass media communication, this systematic review probably overstates the effectiveness of SM.

So what does work?

Let’s take the three behavioural challenges which have a strong emerging evidence base about what works and what doesn’t work: tobacco cessation, road safety, and obesity prevention.

What reduces tobacco smoking?

According to a 2000 US National Cancer Institute study, media campaigns can produce reductions in smoking, “but only when the rest of the social structure actively changes the environment of the smoker.” [viii]

A 2001 World Health Organisation review of anti-smoking campaigns from 9 countries and 6 US states and concluded that media campaigns can work when combined with counseling services, price increases, advertising bans and indoor smoking bans, and plenty of news stories. [ix]

A 2004 review concluded “substantial evidence indicates that higher taxes and clean air laws can have a large impact on smoking rates. Evidence also indicates that media campaigns when implemented with other policies are important.” [x]

What reduces road accidents?

The World Health Organisation’s 2004 “World Report on Traffic Injury Prevention”, an authoritative global review of road safety interventions, does not mention SM, but  notes “when used in support of legislation and law enforcement, publicity and information can create shared social norms for safety. However, when used in isolation, education, information and publicity do not generally deliver tangible and sustained reductions in deaths and serious injuries.” [xi]

A 2004 systematic review into the effectiveness of anti-drink-driving programs concluded that mass media campaigns that are carefully planned and well executed, that reach a sufficiently large audience, and that are implemented together with other prevention activities – such as highly-visible enforcement – are effective in reducing alcohol-impaired driving and alcohol-related crashes.[xii]

Summarising the evidence, Woolley (2001) [xiii] concluded that mass media advertising, when used alone, is unlikely to bring about significant road user behaviour change. However, advertising was found to play an important role in supporting other road safety activities, in particular enforcement.

Barry Elliott, a Australian researcher who carried out a systematic review of road safety campaigns, summed it up pithily: “you can’t sell road safety like soap.” [xiv]

What reduces obesity?

A recent US National Research Council report, Local Government Actions to Prevent Obesity provided a nice summary of the kinds of interventions that have the greatest potential to tackle childhood obesity. According to the press release: “Many of these steps focus on increasing access to healthy foods and opportunities for active play and exercise.  They include providing incentives to lure grocery stores to underserved neighborhoods; eliminating outdoor ads for high-calorie, low-nutrient foods and drinks near schools; requiring calorie and other nutritional information on restaurant menus; implementing local “Safe Routes to School” programs; regulating minimum play space and time in child care programs; rerouting buses or developing other transportation strategies that ensure people can get to grocery stores; and using building codes to ensure facilities have working water fountains.”

In other words, if we wanted to run a comprehensive anti-obesity program then the skill mix would include an incentives manager, a regulator, a building code planner, a nutritionist, a transport planner, an educator (and a courageous politician or two to drive these changes) but not a marketer.

So what, exactly, is wrong with social marketing?

Social Marketing is a system of practice that does many things well. The problem is it what does not do well.

1) Just following orders

SM almost invariably assumes the prescribed behaviour or action is right, just, appropriate, and do-able. SM rarely goes behind the funding agency’s brief, so we have:

“Just think.” (the AFL’s anti-alcohol-violence campaign);

“Quit now before it’s too late” (Australian Gov
ernment’s tobacco campaign)

“Slow down stupid.” (Queensland’s anti-speeding campaign).

SM takes it as given that the particular behaviour should be adopted and can be adopted. It does not ask whether the prescribed behaviour make sense, whether it is capable of being adopted or whether it needs to be reinvented, matured, debugged, or replaced with an entirely different behaviour.

For instance, California’s anti-drug campaign has now abandoned the typical “Just don’t do it” or “Talk to your kids” approaches and opted for a far more subtle “Dinner makes the difference” approach, where the behaviour is simply to have dinner with your kids. This requires a fundamental re-think of the problem and the solution. We simply do not see this in typical SM programs where the funding agency’s assumptions are rarely challenged.

(The reason, of course, is the structural separation, in separate silos, of the policy-bods and boffins who devise strategies, the health workers who implement them, and the educators and marketers who communicate them.)

2) Context blindness

SM and CBSM are tokenistic in their treatment of context. Context, as we discussed, is central to the adoptability of behaviours and products. It’s more than the usual cursory consideration of the 4 Ps: “product, price, place, promotion”. Instead the entire contextual system needs to be the subject of strategizing and modification, including physical infrastructure, service design, place design, management and regulatory systems. Getting these right is usually what makes or breaks a change program, as we’ve seen in tobacco control, road safety, pollution control and littering.

This work can only be done by multi-disciplinary teams using a system-based approach. Again, it’s easy to see how silos enforce dysfunction here, and busting or bypassing silos is the prerequisite for effective systemic interventions.

(By the way, this is not nearly as hard as it sounds. For a rapid method for identifying doable interventions in a whole system, see How to make a theory of change.

3) Crop spraying

SM, as almost universally understood and practiced by governments, is all about big budget mass media advertising. This approach treats people as isolated individuals and sprays them from afar with messages the same way a crop duster sprays a crop of canola. But who still thinks that human societies change this way?

Fifty years of Diffusion of Innovations scholarship and more recent social network studies (notably the remarkable work Nicolas Christakis and James Fowler on the diffusion of obesity, happiness and smoking cessation through social networks) demonstrate that decisions to adopt new behaviours travel primarily along social networks of people who know and respect each other, on a wave of conversations, and mass media has very little to do with it.[xv]

The programs that are likely to influence voluntary behaviour change are therefore those based on fine-grained, conversational, local approaches (like facilitated workshops, forums, field days and the like). Unfortunately, the advertising agencies that win big budget SM campaigns have no incentive to share this insight with their funders.

4) Theory fetish

It’s a fine thing to have our thinking expanded by psychological and change theories, but it’s another thing to arbitrary impose a particular psychological theory on a real life behaviours of real people leading complicated lives in the real world. It’s quite common to see social marketing and health promotion programs introduced with a statement that “this program is based on the Transtheoretical Model” or the Health Belief Model or Social Learning Theory, or whatever. Excuse me, but this is crazy. The theory of change that informs a program should come from one place only – the reality of people’s lives, and it will be very different for each set and each setting and each moment in time. Generic theories and models can help us “see” better as change agents, but only by getting to know people face-to-face and listening intently to their stories can we begin to construct solutions to their needs.

Craig Lefebvre, a perceptive internal critic of Social Marketing, is clear on this when he writes that “One principle that distinguishes the best social marketers, I believe, is an unrelenting understanding, empathy and advocacy of the perspective of our priority population or community that is not slanted by what the theory or research evidence does or does not tell us.” [xvi]

5) Power blindness

SM and CBSM campaigns tend to be one-sided exercises in power by government-employed professionals who decide what behaviours are wrong, what behaviours are right, who needs to change, and what they need to know. Only problem is: people HATE being given advice by strangers about how they should behave. SM and CBSM don’t even begin to have answers for the waves of denial and resistance that are evoked by well meaning attempts to tell people how they should live their lives. See, for instance, the literature on psychological reactance [xvii] and the Boomerang Effect. [xviii]

Many SM programs have figured out a way to remain oblivious to denial and resistance: they evaluate their efforts at the level of awareness. Awareness, however, cuts both ways. Awareness may help drive change, but it is just as implicated in driving people to do the opposite to what they are told. There’s plenty of evidence, for instance, that marketing efforts may reinforce good behaviour amongst those who are already doing the right thing, but drive greater denial and/or resistance amongst the actual target audience.[xix] Even a seemingly benign effort like asking householders to calculate their ecological footprints has been shown to produce this effect. [xx]

6) Message Fetish

Lastly, SM and CBSM have “message fetish” embedded deep in their genomes. Marketing has always been an art of mass communication. It is concerned, above all else, with language and image. It will always be, for better or worse, about the magic of the message. It’s hopelessly infected with the assumption that the right form of words is the key to the human psyche. If it was that easy we’d all long ago have been living in paradise (or, more likely, hell). It just ain’t that way.

And my point is…

I don’t discount the utility of SM, CBSM and COMBI as communication practices, but as social change practices they fall short. The halo of omnipotence that currently surrounds them is unwarranted. They are a valuable support practices, not the messiah.

There is nothing wrong with marketers being involved in designing change programs. They bring a valuable set of skills and perspectives. In fact a change program that doesn’t involve marketers is probably only slightly less problematic than one that is run entirely by marketers.

However, the ability to change the world will never the shining glory of any one discipline. Successful change efforts happen when engineers, planners, politicians, regulators, facilitators and marketers step out of their cosy professional fugs, mix it up with each other, let their assumptions be challenged, be prepared to defend those assumptions with evidence, and invite the public to genuinely collaborate in this process. That’s when the shining glory begins.

If not SM, then what? I don’t think the alternative is rocket science, just a little uncomfortable:

1) Get the “who” right first

Bypass silos, work in multi-disciplinary teams, and invite the users to share the big decisions with you.

2) Get inspired by what works elsewhere. Don’t start till you’ve got lost in Google and Google Scholar a few times and been genuinely excited by the methods others have used, no matter how unfamiliar.

3) Listen to users and non-users and don’t stop listening till you’ve been startled or confronted by what you hear.

4) Notice your own power and actively share it around, especially with those whose behaviour you hope to change.

5) Think in terms of systems. Map the system and don’t limit your palette of interventions.

6) Get all those who can make a difference around the table before you start planning. Let them share the thinking, the planning and the credit.

7) Intervene in the context. Act to modify the environments in which people make their decisions, and then use communications to draw peoples’ attention to those changes and model appropriate behaviours.

8) Be ready to abandon your own assumptions, even the ones you don’t know you have.

What would an effective process for designing a social change program look like? I’m done my best to evolve one over the last few years. It’s available on my website, see The Enabling Change process.

v2.2 Les Robinson, December 2010 



For more detailed critique of Social Marketing, see:


Tilbury, D., Coleman, V., Jones, A., MacMaster, K. (2005) A National Review of Environmental Education and its Contribution to Sustainability in Australia: Community Education. Canberra: Australian Government Department for the Environment and Heritage and Australian Research Institute in Education for Sustainability (ARIES), pp17





[ii] For a clear expression of rigorous SM practice you probably can’t do better than the work of Craig Lefebvre of The George Washington University School of Public Health and Health Services. He has a thoughtful and well informed blog, of which this summary of his social marketing practice is typical. http://socialmarketing.blogs.com/r_craiig_lefebvres_social/2008/09/planning-a-social-marketing-program.html

[iii] Dr Doug McKenzie-Mohr’s popular Community-Based Social Marketing (CBSM) is a variation of SM that brings in techniques drawn from the social psychology of selling (especially via the work of Robert Cialdini).



[v] Andreason, A. (1995) Marketing Social Change: Changing Behaviours to Promote Health, Social Development, and the Environment, Jossey-Bass, San Francisco, p7

[vi] Stead, M. Gordon, R. Angus, K. and McDermott, L. (2007) A systematic review of social marketing effectiveness, Health Education 107(2) pp126-191


[vii] p180. Oddly only six of the 54 programs relied entirely on typical marketing methods, the rest included methods that no one would define as “marketing”, including counselling, smoking cessation groups, community organization, peer education, classroom lessons, training, citizen taskforces, buddy support, summer camps, exercise classes, and construction of walking paths. Only two, however, used any structural or regulatory methods, which is the point.

[viii] National Cancer Institute (2000) Population Based Smoking Cessation: Proceedings of a Conference on What Works to Influence Cessation in the General Population, Smoking and Tobacco Control Monograph No 12, Bethesda, MD, US Department of Health and Human Services, p200

[ix] Schar, E.H., and Gutierrez, K.K. (2001) Smoking Cessation Media Campaigns From Around the World, Recommendations From Lessons Learned, World Health Organisation, Copenhagen

[x] David T. Levy, Frank Chaloupka, and Joseph Gitchell (2004) The Effects of Tobacco Control Policies on Smoking Rates: A Tobacco Control Scorecard, Journal of Public Health Management Practice, 2004, 10(4), 338–353

[xi] Peden, M. et al (eds) (2004) The World Report on Traffic Injury Prevention, World Health Organisation, p138 http://whqlibdoc.who.int/publications/2004/9241562609.pdf

[xii] Elder R.W. et al (2004) Effectiveness of mass media cam­paigns for reducing drinking and driving and alcohol-involved crashes: a systematic review. American Journal of Preventive Medicine 27(1) p57-65

[xiii] Motor Accident Commission (2001) Road Crash Facts for South Australia, Government of South Australia.

[xiv] Elliott, B. (1993) Road Safety Mass Media Campaigns: A Meta Analysis, Department Of Transport And Communications, Federal Office Of Road Safety

[xv] Social marketer Craig Lefebvre has written a nice series of blogs on putting the “social” into social marketing.


[xvii] “when a communicator tells his audience what conclusion they must draw, there is a significant resistance to attitude change and even a tendency for boomerang attitude change.”  Brehm, J.W. (1966) A Theory of Psychological Reactance, Academic Press, New York 121

[xviii] Ringold, J.R. (2002) Boomerang Effect: In response to public health interventions: Some unintended consequences in the alcoholic beverage market, Journal of Consumer Policy 25 p34-35

[xix] For instance, Leffingwell T.D. et al (2007) Defensively biased responding to risk information among alcohol-using college students, Addictive Behaviours 32 pp158-165