The one model worth memorizing
Every marketing strategy eventually answers the same question: what do we want people to do? Buy. Subscribe. Switch. Recommend. Renew. The answer is always a behavior.
But you can't change behavior directly. You have to change what people believe first. Everything else — positioning, messaging, creative, media — is downstream of one decision: what belief, if changed, would produce the behavior we need?
This is the belief–behavior chain. It's the load-bearing wall of communication strategy. The idea is old enough to have academic credentials and young enough to still catch people off guard in practice.
The chain
The full sequence runs like this:
Message/Experience → Belief → Behavior → Audience → Money
The chain reads right-to-left. Money sits with people — individuals or companies. Which of those people, if they changed their behavior, would produce the most revenue? That's your audience — and choosing it is already a strategic decision. A product might shift from targeting individual users to targeting teams, or from end users to decision-makers. Not because the messaging failed, but because the audience selection was the bottleneck.
Once you've identified the audience, you need a specific behavior from them (purchase, upgrade, renew, refer). Behavior follows from what people believe about you, your category, and their own situation. And belief changes when someone encounters a message or experience that gives them a reason to update what they think.
Five links. Two strategic decisions that most teams collapse into one. The first is who — which audience. The second is what — what belief shift, through what message. Most marketing planning starts at the second and treats the first as settled, defaulting to whoever they targeted last time.
If you can't fill in every link, you don't have a strategy. You have a mood board with a budget attached.
Where it comes from
The academic backbone is Fishbein and Ajzen's Theory of Reasoned Action (1975), later extended into the Theory of Planned Behavior: people act on what they think is true, valuable, or expected by others. Ajzen's addition — perceived behavioral control, can I actually do this thing? — matters because it acknowledges that belief alone isn't always enough. Someone might believe your product is great and still not switch, because switching feels hard or risky or socially awkward.
The practitioner version — the one I've carried in my head for years and eventually put into the Strategy Deck — strips the academic scaffolding and keeps the operational core: identify the current belief, define the desired belief, design the communication that bridges them. Current belief leads to current behavior. You need a different behavior, so you need a different belief. Communication is the intervention.
So far, this is the orthodox version. Textbook stuff.
Here's where it gets interesting. The arrow also runs the other way.
Leon Festinger called this cognitive dissonance (1957). When people act in ways that conflict with their beliefs, they don't sit comfortably with the contradiction — they update the belief to match the action. Buy an expensive product, then convince yourself it was worth it. Switch to a new tool, then develop loyalty to justify the switch. Start using a product for a trivial reason, then construct a narrative about why it's actually the right choice.
Adam Ferrier, a psychologist-turned-advertising-strategist, built an entire framework on this reversal. His book The Advertising Effect argues that the most effective way to change behavior is through action, not persuasion. Get someone to do something small — try it, sample it, commit to a tiny step — and the belief follows. He calls these "action spurs."
So the chain isn't a one-way street. It's a loop:
Belief → Behavior → Belief (updated) → Behavior (reinforced)
Communication can enter at either point. You can change what people think (belief-first), or you can change what people do (behavior-first), and the other side adjusts. The strategic question isn't which direction is "correct." It's which entry point is available given your audience, your product, and your constraints.
This is the part most teams miss — not the chain itself, but the fact that it has two doors.
The chain in practice
The model becomes useful when you make it specific enough to be wrong. Three examples — different industries, different entry points into the chain, and each one bumping into a different limitation of the model.
Examples
Audience: People who drink water but find the category invisible — especially younger consumers at events, concerts, and social settings where what you hold signals who you are.
Current belief: "Water is boring. It's what you drink when you're not drinking something interesting."
Desired belief: "This water is for people like me. It doesn't compromise my identity."
Desired behavior: Choose Liquid Death over other water brands (or over not buying water at all).
Entry point: Belief-first. The product is water. Nothing changed about the liquid. What changed was the belief about who water is for and what carrying it says about you. Tallboy cans, punk branding, merch drops, a name that sounds like an energy drink — all of it attacked the assumption that water is a category without identity.
Note what happened at the audience level: Liquid Death didn't just change the message to existing water drinkers. They expanded who the audience was by making people who had never thought about water as a choice suddenly see themselves in it. The chain reads right-to-left. But sometimes the biggest move is widening who's in it.
And here's where the model strains: "this brand feels like me" isn't really a belief in any propositional sense. It's identity. The chain handles "I believe this product has feature X" cleanly. It handles "this brand is me" less cleanly. Liquid Death's success was as much about cultural recognition as about belief change. The model points you in the right direction, but the actual mechanism is warmer and messier than the framework suggests.
Audience: Teams using email for internal communication (which was everyone).
Current belief: "Email works fine for team communication. It's how work gets done."
Desired belief: "Email is where work goes to get buried. There's a faster way."
Desired behavior: Start a free Slack workspace with your team.
Entry point: Belief-first, but targeting a belief people didn't know they held. Nobody was actively dissatisfied with email — it was invisible infrastructure, like air. Slack's early messaging didn't compare features. It reframed the status quo: your current tool has a cost you're not measuring. The belief shift was a seed of doubt about what "fine" actually meant.
This is the hardest kind of belief work — dislodging a belief that doesn't feel like a belief at all. It feels like reality. The chain still applies, but the message has to do more work because the current belief is structural, not conscious. And even when Slack succeeded in shifting the belief, organizational inertia often blocked the behavior. Plenty of people became Slack-curious inside companies that had standardized on email and weren't about to change. Belief changed. Behavior didn't — at least not until a decision-maker with purchasing authority also converted. The chain is clean in theory; in organizations, it runs into walls that aren't made of beliefs at all.
Audience: People who drink coffee daily but assume good espresso requires a barista, a real machine, or at least some skill.
Current belief: "You can't get real espresso at home without learning how to make it."
Desired belief: "This is actually good — and I just pressed a button."
Desired behavior: Purchase a Nespresso machine and subscribe to capsules.
Entry point: Behavior-first. Nespresso didn't try to argue people out of their belief about espresso. They designed around it. Capsules showed up in hotel rooms, airport lounges, offices — places where you tried it before anyone asked you to believe anything. The first sip did the work. The belief updated itself after the behavior.
This is Ferrier's loop in practice. The belief ("real espresso can't come from a capsule") was too entrenched to attack with messaging. But once someone had a decent shot from a machine they didn't own, the belief softened on contact with experience. Behavior changed belief. Belief reinforced behavior. Worth noting: this only worked at the trial stage because the moment of decision was still conscious — someone chose to try it. The interesting part is what happened next. Once the capsule became part of the morning routine, belief stopped mattering. Nobody thinks "I believe Nespresso makes good espresso" while half-asleep at 7am. The behavior became habitual, and the chain — which assumes someone is thinking — quietly stopped applying.
The honest limitations
The examples above already hint at where the model runs thin. But it's worth being explicit, because pretending a model explains everything is how you miss the places where it doesn't apply.
Habitual behavior. Some behaviors are automatic, not belief-driven. Disrupting habit requires a different kind of intervention — a context disruption, a life transition, a moment when the autopilot turns off. The belief–behavior chain assumes someone is thinking. A lot of purchasing isn't.
Social and structural constraints. The Theory of Planned Behavior tried to account for this with "perceived behavioral control," but in practice, organizational inertia and social norms are separate forces. Belief is a necessary condition. It's rarely a sufficient one.
Emotional and identity-driven behavior. "I believe this product has feature X" is a different cognitive object from "this brand feels like me." The chain works for the first. It gets blurry with the second.
These limitations don't make the model wrong. They make it incomplete — which is what every honest model should be. If you only have time for one model, this is the one. But hold it lightly.
What I actually use it for
The belief–behavior chain is the first thing I reach for when someone asks me to approve a campaign or review a content plan. Not because it solves everything, but because it turns "it didn't work" into a question with structure:
- Was the audience wrong? (A different group would have been higher-leverage for revenue.)
- Was the belief diagnosis wrong? (Right audience, wrong assumption about what they think.)
- Was the message wrong? (Right belief target, but the communication didn't shift it.)
- Was the behavior link wrong? (The belief shifted but friction, switching costs, or constraints blocked the action.)
Four failure modes. Four different fixes. Most teams treat a failed campaign as a creative problem. Usually it's a diagnostic one.
References: Fishbein, M. & Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior. Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. Festinger, L. (1957). A Theory of Cognitive Dissonance. Ferrier, A. (2014). The Advertising Effect: How to Change Behaviour.
© Ilya Petrov, 2026.