Most social marketing reaches for the same few behaviour change models — stages of change, COM-B, the ones everyone learned first. But the model you know best is rarely the one the job needs. And the wrong model quietly builds the wrong campaign.
Pick up almost any social marketing brief and you'll find a behaviour change model (or often models) already baked in. Usually it's the transtheoretical model — stages of change. It's a good model. We've used it for years and it's shaped some of our best work. But it's become a reflex. The model gets chosen before anyone's worked out what behaviour they're trying to shift, for whom, or why it isn't happening yet. That's backwards. The model should come last, once you understand the challenge — not first, out of habit.
Theory, framework, model and delivery guide aren't the same thing
A lot of the confusion starts here, because these words get used as if they're interchangeable. They're not, and the difference matters.
A theory explains why behaviour happens. A framework gives you a structure to organise your thinking. A model represents how change moves, or what state someone's in. — stages of change is a model, so is COM-B. And a delivery guide is an execution tool: it tells you how to land a message, never what to say or why you're saying it. EAST — easy, attractive, social, timely — is the clearest example of a delivery guide. It's brilliant for sharpening a prompt. It can't tell you what behaviour to prompt, or who to prompt it in.
When you treat a delivery guide as if it were a theory, you skip the thinking that decides whether the campaign will work at all. You end up with a slick, well-timed message pointed at the wrong behaviour — or aimed at the wrong thing altogether: the individual, when the behaviour is largely shaped by the environment around them. Knowing which tool you're holding — and what it can and can't do — is the first discipline.
Most models assume one rational person deciding alone
We think there’s a bigger problem. Most of the models we inherited were built in the “West”, on a particular picture of how people change: a single person, weighing up costs and benefits, moving themselves along from not-doing to doing. Stages of change pictures it. The health belief model pictures it. They're not wrong — but they assume the individual is the unit of change. We're not alone in saying so — there's a growing push within social marketing itself to move past individualistic, Western-built models and to centre Indigenous knowledge as theory in its own right.
For a lot of people, that's just not how decisions get made. When we evaluated Te Whatu Ora's F.A.S.T. Stroke campaign, we found nearly a third of people — 32.7% — made health decisions collectively, alongside others. Māori and Pasifika respondents were far less likely to decide alone. If your model assumes a lone decision-maker and your audience decides as a whānau, the campaign is talking to the wrong person in the room. You can run a textbook stages-of-change campaign and still miss, because the stage that matters isn't in one person's head — it's in the conversation around the kitchen table.
There's a whole tradition that starts from exactly there. Practice theory — developed in sociology and applied directly to gambling harm — says people don't take up a behaviour because they simply decide to. The behaviour is a practice that gets reproduced through them, held together by three things: the materials that make it possible (the machines, the apps, the venues, the phone in your pocket), the competences it takes to do it, and the meanings it carries — mateship, escape, a normal Friday night. Change one of those elements and the practice starts to come apart.
That's a different place to stand. The question stops being how do we move this person along, and becomes what holds this practice in place, and which part of it can we shift. It puts the cause of harm on the practice and the system that reproduces it — the product, the environment, the shared meaning — not on the individual caught inside it. Most social marketing quietly assumes the product is neutral; this way of working doesn't. And most behaviour change models can't take you here, because they were built to move a person, not to loosen a practice.
Different worldviews ask different questions
This isn't only about who decides. It's about what counts as wellbeing in the first place. Western models carry Western assumptions — individual, rational, linear. Frameworks grounded in te ao Māori and in Pasifika knowledge start somewhere else. Te Whare Tapa Whā holds the physical, mental, whānau and spiritual sides of a person as inseparable. Te Pae Māhutonga puts cultural identity and collective autonomy at the centre. The Kakala framework moves the way Pacific relationships and reciprocity actually move. Fonofale roots wellbeing in family, culture and the ties between them, and vā — the relational space between people that must be actively tended — sits underneath how obligation and relationship work across the Pacific.
These aren't cultural decoration to layer over a Western model once the strategy's done. They're complete systems for understanding change, and for the communities carrying the most harm, they're often the ones that read the situation most accurately. We use He Awa Whiria (the braided rivers approach) to bring tāngata whenua knowledge and tāngata Tiriti knowledge together as equals, each strengthening the other. You can see that braid up close here: te ao Māori frames of practice — tikanga, kawa, whakapapa — and Western practice theory's three elements (materials, competences and meanings) reach the same insight from their own foundations, that practices are collective, reproduced and able to be changed. And te ao Māori holds what practice theory has no word for — that a practice has its own whakapapa, binding people through obligation across generations — on its own terms, not as an add-on to a Western model. Neither stream gets translated into the other; both run, and each sharpens the read. That's what we call Tiriti-dynamic social marketing — an approach we've developed and published. We hold to it first because tino rangatiratanga means communities keep authority over the knowledge that explains their own lives — and we keep finding it isn't only the right thing to do. It works better.
So how do we choose?
We start with the behaviour and the people — not the model. What exactly are we trying to shift? For whom? And why isn't it happening already?
None of those questions can be answered from the desk. Understanding a behaviour well enough to choose a model takes deep work — alongside experts from the communities you're trying to reach, in exploratory mahi with the people actually living it, and across community knowledge, the literature and the evidence at once. And it means choosing the model with those communities, not for them. They often already hold the frame that fits; the work is to listen for it.
That last question — why isn't it happening already — is where a model like COM-B earns its place, because it's a diagnostic. It asks whether the gap is capability, opportunity or motivation. On He Wāhi Haumaru Breastfeeding NZ, the barrier wasn't knowledge — parents knew breastfeeding was good for their pēpi (baby). The gap was opportunity and social support: feeding in public felt unwelcome. So we built safe, visible online spaces instead of printing more information. A different model would have sent us the wrong way.
Once you understand the real barrier and how your audience actually decides, the right model — or, more often, the right combination — becomes clear. Sometimes it's stages of change, used honestly — a map of where people are, not a treadmill they walk along: most people experiencing harm aren't contemplating anything, so you meet them gently where they are rather than waving a helpline at them. Our own early gambling work (Choice not chance) leaned hard on individual choice. The programme has moved past that and so have we — and practice theory is what names the shift: harm sits on the product and the system that reproduce the practice, not on the person caught inside it. And often the model that fits isn't on the Western shelf at all — it's a kaupapa Māori or Pasifika frame the community already holds, and the work is to back it, not to translate it into something more familiar.
A behaviour change model is a tool, not an answer. Reach for it too early and it quietly makes decisions you haven't thought through. Choose it for the job in front of you and the people you're actually trying to reach — and braid it with what those communities already know, holding to the ethics that knowledge carries: manaakitanga, and respect for the mana of the people it belongs to. Then it does what it's meant to: help you see clearly.
Want to talk about choosing the right approach for your kaupapa? Let's kōrero.
.svg)
.svg)
.svg)
