Transtheoretical Model: An S-Tier Behavioral Designer’s Guide to Stages of Change

A remarkable number of behavior-change products fail in the same place: the designer assumed the user had already decided to change, when the user had not even admitted there was a problem. Precontemplation, the opening stage of the Transtheoretical Model, is where about forty percent of American smokers sat in three large representative surveys, and it is the stage almost no onboarding flow, habit app, corporate training module, or government health campaign is designed for. The Prochaska-DiClemente model tells you what that crowd looks like, what they need, and what they will do to any Action-stage tool you inflict on them.

The Transtheoretical Model (TTM), also known as the Stages of Change model, is one of the two most-used theories in health-behavior research, and that is not an accident. When Painter and colleagues coded a random sample of 2000–2005 articles from ten leading health journals, it tied with Social Cognitive Theory at the top, used by 27.5% of the papers that named a theory. It gives you five stages a person moves through when changing any durable behavior — Precontemplation, Contemplation, Preparation, Action, Maintenance — plus a toolbox of ten Processes of Change that accelerate movement between them, a self-efficacy dimension borrowed from Bandura, and a decisional-balance ledger borrowed from Janis and Mann. Read charitably, it is the closest thing psychology has to a complete operating system for human change. Read cynically, it is a stage taxonomy whose headline promise, that matching help to the stage multiplies results, has not survived the large trials. The truth is somewhere in the middle, and the middle is where the design opportunity lives.

This is the S-Tier designer’s guide to Prochaska and DiClemente’s framework. By the end you will know what the five stages actually predict, which processes work in which stage, the three fair criticisms the literature has sustained for four decades, how the stages map almost one-for-one onto the four Experience Phases of the Octalysis Framework with a crisp Precontemplation lobby the standard phases quietly assume away, and the concrete design moves I use to avoid the Action-stage-only trap that so many shipped habit products fall into.

Speed Run Notes

• Five stages, one real idea. Precontemplation, Contemplation, Preparation, Action, Maintenance. The point is stage-matched help: what moves a person depends on where they stand.
• 40/40/20 is a smoker statistic. Three big US smoker surveys split 40% Precontemplation, 40% Contemplation, 20% Preparation. Exercise and other countries look very different, so measure your own.
• Ten Processes of Change are the verbs. Behavioral ones like stimulus control peak in Action and Maintenance for every behavior studied; the early-stage pattern varies by behavior.
• Decisional balance is the best-replicated signal. Across 48 behaviors, reaching Action meant about a 1.00 SD rise in perceived pros and a 0.56 SD drop in perceived cons.
• The weak spot is tailoring. A Cochrane review of 41 trials found stage-based smoking help no better than the same help unstaged. Use stages to diagnose, not as a guarantee.
• Relapse is normal, and rarely total. Only about 15% of relapsers fall back to Precontemplation. Build a re-entry door that says you are not starting over.

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003 — years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

What Is the Transtheoretical Model

The Transtheoretical Model — TTM, also called the Stages of Change model — is an integrative behavior-change framework developed by psychologists James O. Prochaska and Carlo C. DiClemente at the University of Rhode Island between 1977 and 1983. The “transtheoretical” in the name is a promise the authors made to themselves: the model would not be loyal to any single school of psychotherapy but would instead identify the processes of change that work across schools, synthesise them, and sequence them against a common staircase of behavior change. In practice the framework has four moving parts: the five stages, the ten processes, self-efficacy, and decisional balance.

The stages are the most recognisable piece. Precontemplation is the stage before a person has any intention to change a given behavior in the next six months. They may not perceive the behavior as a problem, or they may have tried and given up often enough that further attempts feel pointless. Contemplation begins the moment the person acknowledges the problem and starts seriously weighing change — typically within the next six months but not imminently. Preparation is the thirty-day runway: the person intends to act soon, is gathering information, experimenting with small adjustments, and assembling the resources the change will require. Action is the visible behavior change itself, lasting up to six months of sustained new behavior. Maintenance begins after six months of consistent action and can last years; the behavior has become stable but relapse is still a meaningful risk. Some later formulations add a sixth stage, Termination, for the rare state in which relapse is not considered a risk at all — most applied work quietly drops it.

The ten Processes of Change are the verbs the model gives you for moving people up the staircase. Five are cognitive-experiential — consciousness raising, dramatic relief, self-reevaluation, environmental reevaluation, and social liberation — and five are behavioral — self-liberation, reinforcement management, helping relationships, counter-conditioning, and stimulus control. Prochaska and DiClemente’s biggest empirical contribution was not the staircase itself but the discovery that the processes are stage-matched: cognitive-experiential processes dominate early, behavioral processes dominate late, and applying the wrong process to the wrong stage reliably produces stalled change.

The two secondary constructs tie the picture together. Self-efficacy, borrowed from Albert Bandura, tracks the person’s confidence that they can sustain the new behavior against specific temptations; self-efficacy climbs steadily across the stages. Decisional balance, borrowed from Irving Janis and Leon Mann, tracks the subjective weight the person assigns to the pros and cons of change; the ratio inverts somewhere in Contemplation and continues to move until Maintenance. Together these two instruments produce a quantitative fingerprint of where the user sits on the staircase, which is the operational hook every practical application of TTM depends on.

The Core Findings of Prochaska & DiClemente

The core empirical findings that made TTM a durable framework rather than a pretty diagram are worth naming precisely, because each of them carries design implications that most practitioners skip.

Finding 1: People do not go straight from “not thinking about it” to “doing it”

The stage model grew out of smokers who were quitting on their own, and its first big empirical paper is the 1983 one, not the 1982 one. The 1982 article was the theoretical case for a transtheoretical therapy. In 1983 Prochaska and DiClemente studied 872 adults trying to quit without formal treatment and sorted them into five groups: precontemplation, contemplation, action, maintenance, and relapse. There was no Preparation stage yet. DiClemente, Prochaska and colleagues carved it out of Contemplation in 1991, in a sample of 1,466 smokers, for people intending to quit within the next thirty days.

The finding that mattered in 1983 was that each group leaned on a different mix of processes. Precontemplators used the fewest change processes of anyone, while counterconditioning and stimulus control were used most in action and maintenance. Relapsers looked like a blend of contemplators and people in action, which is the first hint of the spiral. Any policy or product that treats its audience as a homogeneous pool of “people trying to change” is addressing a mixture as if it were one thing.

Finding 2: The 40/40/20 split is real for American smokers, not for everyone

The famous shape comes from Velicer and colleagues (1995), who staged three large samples of smokers: 4,144 from a random-digit-dial survey in Rhode Island, 9,534 from a statewide survey in California, and 4,785 smokers from 114 worksites. All three came out at about 40% Precontemplation, 40% Contemplation and 20% Preparation, and the split held across age groups below 65. The authors drew the conclusion that still matters: most smokers sat in the first two stages while most programs were built for the third.

That is a strong, replicated result about American smokers in the early 1990s. It is not a law of behavior. The Cancer Prevention Research Center’s own overview of the model reports European smoking samples at roughly 70% Precontemplation, 20% Contemplation and 10% Preparation. Marshall and Biddle’s 2001 meta-analysis of physical activity, 91 samples and more than 68,000 people, found almost the mirror image: 14% Precontemplation and 36% already in Maintenance. Their review also warns that the distribution depends on how a study samples and recruits. The design implication survives in a more useful form. Your funnel almost certainly holds a mix of stages, and the only way to know your mix is to measure it in your own audience.

Finding 3: Tailored help beat a standard manual in the trial that made the model famous

The result that launched a generation of stage-matched programs is Prochaska, DiClemente, Velicer and Rossi (1993). They randomized 756 smokers, balanced by stage, to one of four arms: the American Lung Association’s standardized self-help manuals, stage-matched manuals, stage-matched manuals plus interactive computer-generated feedback reports, or that package plus counselor calls. At 18 months the interactive expert-system arm had more than double the abstinence rate of the standardized manuals, and it was the best or tied for best at every stage.

Two details matter for anyone quoting it. The doubling is the interactive expert-system arm against a standard manual, not “stage matching” against everything else; at 12 months the stage-matched manuals on their own were no better than the standard ones. And the larger literature that followed did not hold the line, which is Critique 3 below. The mechanism the authors proposed is still sensible, since a precontemplator handed a quit plan has nothing to do with it and a smoker who is ready to act gets little from another awareness pamphlet. Treat the 1993 trial as the strongest early evidence for tailoring, not as proof that matching doubles results.

Finding 4: Decisional balance and self-efficacy are the measurable stage signals

Decisional balance is the most robust number in the model. In 1994 Prochaska and colleagues measured the pros and cons of changing across 12 problem behaviors and found the same pattern of change in decisional balance across all twelve: the cons of changing loom larger than the pros in Precontemplation, and the balance tips toward the pros as people approach Action. Hall and Rossi’s 2008 meta-analysis took the test to 120 datasets, 48 health behaviors and nearly 50,000 people in 10 countries, and pinned down the size. Moving from Precontemplation to Action corresponds to a 1.00 standard deviation increase in the perceived pros and a 0.56 standard deviation decrease in the perceived cons.

That asymmetry is a design instruction. The pros side has to move about twice as far as the cons side, which is why a frictionless sign-up flow on its own rarely wakes up a precontemplator; removing friction works on the smaller lever. Self-efficacy tends to climb across the stages as well. Both can be read with a question or two inside a product, a confidence item and a pros-versus-cons prompt, which makes them practical triggers for a stage-appropriate nudge. Note what the data does and does not say: these are cross-sectional differences between people at different stages, so they tell you where a user probably stands, not that a rising score will cause the next move.

Finding 5: Relapse cycles the user, it does not end them

Once action is taken, relapse is the rule rather than the exception for most health behaviors, and Prochaska, DiClemente and Norcross (1992) describe people as typically recycling through the stages several times before the change holds. The part that matters for design is where relapsers land. The Cancer Prevention Research Center’s summary of the smoking and exercise data puts it at only about 15% regressing all the way to Precontemplation; most go back to Contemplation or Preparation and begin planning the next attempt. That is why Prochaska drew the staircase as a spiral. A design that treats the first relapse as a terminal event imposes a linear model on a spiral reality and loses the very users the model expects to try again.

What Prochaska & DiClemente Got Right

More than four decades of empirical work let us judge TTM with a harder standard than “is this useful?”. By that standard, three pieces of the original framework look, on the evidence, to be genuinely right.

The first is the population-level design claim. The insistence that a change target is a heterogeneous mixture of stages rather than a homogeneous pool of “people ready to change” holds up in every population anyone has staged, even though the mix itself varies enormously from smokers to exercisers to countries. It is also the framing that most often unsticks an engagement program. Recount the users by stage instead of by demographic, and the usual leak shows itself: the content and the incentive structure both assume an Action-stage audience that turns out to be a minority of the inbound traffic.

The second is the stage–process matching rule, and it needs a more careful statement than it usually gets. Rosen’s 2000 meta-analysis pooled 47 studies across smoking, substance use, exercise, diet, psychotherapy and other problems. Behavioral processes peaked in Action and Maintenance for every problem studied, and self-liberation and counterconditioning showed the largest differences between stages, with effect sizes around d = 1.0. But stage explained only about 11% of the variance in cognitive-affective process use and 14% in behavioral process use, and no single trajectory held across all behaviors. So the late-stage half of the rule is solid, and the early-stage half depends on the behavior.

That is still more actionable than most behavior-change theories manage. It tells you that a user who is already acting needs environment design, replacement routines and reinforcement rather than more persuasion, and it warns you to check the early-stage pattern for your own behavior instead of copying the smoking literature.

The third is the humane treatment of relapse. Prochaska and DiClemente named the spiral pattern before the public-health community was comfortable with it, and they kept naming it until it stuck. Treating relapse as a normal station on the line rather than as a terminal failure changes the emotional tone of the entire intervention, the rhetoric of the user interface, and, most practically, whether a returning user is treated as a cost or as someone the model expects to try again. The spiral framing is Prochaska’s most important gift to design ethics.

A fourth claim — the integrative synthesis of processes across schools of therapy — is the one the original name was built around and the one most outside their immediate field remember least. By extracting a common process vocabulary from behaviorism, psychoanalysis, cognitive therapy, humanistic therapy, and existential therapy, the authors gave the field a Rosetta Stone. You can argue about the fidelity of the translation, but you cannot argue that the Rosetta Stone was not needed. It was, and it still is.

Where the Transtheoretical Model Falls Apart

The honest review of TTM also names three places the model is weaker than its fans claim, and one of them has consequences that every practical application needs to design around.

Critique 1: The stages may be descriptive clusters, not causal states

The most substantial academic critique — associated with Robert West and colleagues — argues that TTM’s stages are reifications of descriptive clusters rather than true causal states with bright lines between them. On this reading, assigning a user to “Contemplation” versus “Preparation” is less like assigning them to a discrete country on a map and more like arguing about where the beach ends and the ocean begins. The critique matters because a stage without a bright line cannot, in principle, support a crisp stage-matched intervention; the interventions that work may in fact be continuous functions of readiness rather than discrete stage-triggered scripts. The data are compatible with both readings, and the design implication is that staging should be treated as a useful heuristic, not as an engineering spec.

Critique 2: The six-month and thirty-day thresholds are arbitrary

A second fair critique is that the specific time horizons that distinguish Precontemplation from Contemplation (six months) and Contemplation from Preparation (thirty days) are not derived from data. They were the authors’ operational convention, and decades of use have made them feel canonical, but they were never validated against a population-specific finding that behavior change has a six-month anticipatory horizon. For some behaviors — quitting a decades-long smoking habit — the six-month horizon is probably conservative; for others — adopting a new software workflow — six months is roughly eternity. Applying the canonical thresholds without adjusting for the target behavior is one of the ways TTM-based interventions quietly fail.

Critique 3: Large randomised trials have produced modest effect sizes

The most painful finding in TTM research is that the stage-matched advantage, which looked strong in early trials like the 1993 expert-system study, shrinks or disappears in the large controlled literature. Bridle and colleagues’ 2005 systematic review found only limited evidence that stage-based interventions changed health behavior. The Cochrane review by Cahill, Lancaster and Green pooled 41 trials and more than 33,000 smokers. In the four trials that compared a stage-based and a standard version of the same intervention head to head, the stage-based version was neither more nor less effective, with risk ratios of 0.93 for self-help and 1.00 for counselling. Stage-based self-help and counselling still beat no intervention, so the materials work; what the reviewers could not find was added value from tailoring them to stage.

Physical activity tells the same story from a different corpus. Romain and colleagues’ meta-analysis of 33 randomized trials and 10,350 participants put stage-matched programs at d = 0.36 and non-matched TTM programs at d = 0.23, a difference that was not statistically significant. What did predict efficacy was how many TTM constructs a program used, especially self-efficacy and the processes of change. The charitable reading is that real-world implementations rarely achieve the tightness of matching the model requires. The less charitable reading is that the stage label is the least useful part of the model. Either way, the takeaway for a product manager who has to justify a TTM-based build is concrete: invest in the processes and the confidence-building, and treat the stage as a routing hint.

The Brain on Stages of Change

TTM is a behavioral framework, not a neuroscientific one, and the honest place to start is that the five stages have not been imaged. Studies that sort people by TTM stage and compare their brains are rare enough that the stage-by-stage brain maps circulating in design writing are extrapolation. What does exist is narrower, and more useful, because one of its findings cuts against the popular story.

Precontemplation has no imaging literature of its own. The design reading that it is a salience problem rather than a stubbornness problem is a reasonable inference from the behavioral data, since precontemplators use fewer change processes than anyone else, and it should be held as an inference.

Contemplation is where the popular story says the brain visibly weighs pros against cons. The closest direct test went looking for exactly that conflict. Wilson, Creswell, Sayette and Fiez (2013) measured how ambivalent 82 quit-motivated smokers felt about smoking, then scanned them as they held and looked at a cigarette. Ambivalence was not related to activity in conflict-processing regions at all. Instead, the more ambivalent a smoker was, the weaker the cue response in reward and attention areas, including the rostral anterior cingulate, medial prefrontal cortex and caudate. For design, that reframes ambivalence: it may be a sign that the old behavior’s pull is already fading, which is a reason to support the contemplator rather than push harder.

Preparation is where implementation intentions do their work, and here there is real imaging evidence, though about planning rather than about TTM. Gilbert, Gollwitzer, Cohen, Oettingen and Burgess (2009) compared intentions tied to a specific cue with intentions people had to initiate on their own. Cue-linked intentions went with more activity in medial rostral prefrontal cortex (medial Brodmann area 10) and better performance; self-initiated intentions leaned on lateral area 10 within a frontoparietal network and were carried out less reliably. Setting a quit date or writing an if-then plan is the behavioral version of handing the work to the more reliable system before Action arrives.

Action and Maintenance line up with the habit literature. Tricomi, Balleine and O’Doherty (2009) gave people extended training on a rewarded task and found activity growing in the right posterior putamen, within the dorsolateral striatum, as responding became habitual. That is consistent with Maintenance users describing the behavior as “just what I do now,” but it is a finding about habit formation, not a measured property of the Maintenance stage. Designing for Maintenance still means designing for handoff: stable cues, a stable context, and a product that the user needs less over time.

TTM vs Other Motivation Theories

The Transtheoretical Model sits in a neighbourhood of frameworks, and understanding which one to use when saves a lot of wasted effort.

TTM vs the Health Belief Model. The Health Belief Model, developed by US Public Health Service researchers in the 1950s, with self-efficacy added in the 1980s, explains the decision to act in terms of perceived susceptibility, severity, benefits, barriers, cues to action, and self-efficacy. It is strong at predicting the single decision and weak at predicting the trajectory of sustained change. TTM picks up where the Health Belief Model leaves off: once the person has decided, what happens next, and how do we keep them moving? Practitioners routinely use the two in sequence.

TTM vs the Theory of Planned Behavior. Ajzen’s Theory of Planned Behavior models intention as a function of attitudes, subjective norms, and perceived behavioral control, and models behavior as a function of intention plus perceived behavioral control. It is sharper than TTM on what forms the intention but weaker on what happens in the gap between intention and sustained action. The two frameworks are complementary: TPB predicts the arrival at Preparation; TTM predicts what to do after.

TTM vs Self-Determination Theory. Self-Determination Theory, developed by Deci and Ryan, models motivation as a continuum from external regulation through to intrinsic motivation, gated by the three basic psychological needs of autonomy, competence, and relatedness. SDT is stronger than TTM on why behavior change sticks once it sticks — because the motivation has been internalised — but weaker on the staged mechanics of getting there. The two frameworks are natural partners: SDT tells you what kind of motivation to aim for at each stage; TTM tells you which processes to deploy along the way.

TTM vs BJ Fogg’s Behavior Model. Fogg’s model treats a behavior as happening when Motivation, Ability and a Prompt converge at the same moment, B = MAP. Earlier versions called the prompt a trigger, and older write-ups still say B = MAT. It is sharper than TTM on the micro-engineering of a single action but silent on the weeks and months before and after. Fogg is the right lens when you are designing a specific prompt; TTM is the right lens when you are designing the long arc of a change program that may span a year.

TTM vs Dual Process Theory. Kahneman’s System 1 / System 2 distinction, which I covered in a separate S-Tier guide, describes the moment-to-moment cognitive architecture. TTM lives at a higher level of abstraction — the week-to-month progression — but the two interact neatly. Early stages are dominated by System 2 weighing; late stages by System 1 automaticity. Maintenance is literally the stage in which the behavior has become a System 1 default.

The Transtheoretical Model in the Real World

TTM has been applied to almost any behavior you can imagine over the last four decades. The domains where the framework has earned the most respect — and where design teams can still extract unusual leverage in 2026 — are worth looking at in sequence.

TTM in Healthcare

Healthcare is the native habitat of TTM and still the domain in which it does the most measurable work. The classic applications are smoking cessation, weight management, alcohol and substance-use recovery, mammography adherence, sun-protection behavior, and physical-activity adoption. Its closest clinical partner is motivational interviewing, which was built for exactly the ambivalent, Contemplation-stage client the model describes, and the two are routinely taught together. The more recent applications that matter for designers are medication adherence in chronic disease — where a newly diagnosed patient may not yet accept that there is anything to adhere to — and post-operative rehabilitation, where Maintenance is precisely where the typical patient drops out.

TTM in the Workplace

Corporate applications of TTM are more recent and more unfinished. The candidate domains are wellness programs, safety-culture interventions (where precontemplating managers are often the bottleneck), and organisational-change management, where Prochaska’s spiral is a better mental model of how a workforce absorbs a new process than the one-way arrow of a classic unfreeze, change, refreeze diagram. The evidence base at work is thin next to the health literature, and given what Critique 3 found in health, stage matching on the job should be run as a design hypothesis to test rather than sold as a proven lever.

TTM in Education and Learning Design

Education is a fascinating case because the traditional course structure ignores stages entirely — a student enrolled in a curriculum is treated as uniformly Action-ready, whether or not they actually are. Stage-matched course design, by contrast, has shown up in adult-learning settings (where Precontemplation learners are the norm, not the exception), in continuing medical education, and in large-scale behavior-change curricula — notably the computer-tailored programs that grew out of ProChange Behavior Solutions, the company Prochaska founded. The emerging opportunity in 2026 is in AI-tutored learning, where a model that can detect the learner’s stage can route them to Consciousness-Raising content instead of drilling them on Action-stage exercises they are not ready to do.

TTM in Marketing and UX

Marketing and UX are the most commercially visible TTM domains, and also the ones most likely to be doing it badly. Every conversion funnel is a de facto stage model — Awareness, Interest, Desire, Action is TTM’s cousin — but very few marketing teams actually instrument stage detection and very few UX teams design explicit Precontemplation content. That gap is cheap to test. On the UX side, the test is onboarding flows that branch on detected readiness (“are you here to explore, or to commit?”) run against a linear flow that assumes every new user is in Preparation. The commercial version of TTM is less a theory than a permission slip to stop pretending every user is identical.

The Elephant in the Room

The uncomfortable truth about the Transtheoretical Model in 2026 is that most teams who say they are using it are using one-third of it. They have adopted the five-stage picture because it is memorable and fits on a slide. They have not adopted the ten Processes of Change, which are where the actual mechanism lives. They have not instrumented self-efficacy or decisional balance, which are where detection becomes possible. And they have not designed for the spiral, which is where the humane ethics of the framework sits. Reduced to a pretty wheel, TTM is a conference poster. Deployed in full, it is a design system. The gap between the two is where almost every shipped behavior-change product leaks value.

The second elephant is that the criticisms of TTM are fair. A serious designer cannot ignore that large randomised trials have not produced the headline effect sizes the early enthusiasm promised, and cannot pretend the stage boundaries are sharper than the data supports. The honest position is to use TTM as a diagnostic lens that improves your hit-rate, not as a law of physics that guarantees an outcome. Calibrate the expected lift accordingly. The framework earns its place even at the modest effect sizes the careful literature reports, because the alternative — a blanket Action-stage product sprayed at a fully mixed-stage audience — is worse on both efficacy and ethics.

The third elephant is that digital products in 2026 have the tools to instrument TTM properly for the first time in its history, and most of them are not doing it. Passive-sensing data, conversational AI interfaces, and in-product micro-surveys can detect stage indicators that the pencil-and-paper questionnaires of the early 1980s could only dream about. A Contemplation signal in a food-tracking app, a Precontemplation signal in a corporate e-learning module, a Maintenance signal in a financial-wellness product — all of these are technically achievable today. The products that do this work will have a structural advantage over the products that do not. That is the most valuable design opportunity the framework offers right now, and it is the one the field keeps failing to take.

How to Apply the Transtheoretical Model with the Octalysis Framework

The practical payoff of TTM for a designer in 2026 is not in reading the original papers — it is in bolting the five stages onto the four Experience Phases of Octalysis so the whole product speaks the same language. That integration is what this section walks through.

The integration I use when I walk senior product teams through this has four moves.

Move 1: Add a Precontemplation lobby before Discovery

Octalysis Level 2 gives you four Experience Phases — Discovery, Onboarding, Scaffolding, Endgame — and they have served the framework well since Actionable Gamification laid them out in 2015. TTM forces you to notice that the canonical four phases quietly assume the user has already decided to enter. The Precontemplation lobby is the room before Discovery, and TTM insists you design it. In practice this lobby is where you run consciousness-raising content (personal-relevance stories, salient statistics framed to raise the stakes), Core Drive 1: Epic Meaning & Calling assets that tell the user why the behavior matters beyond themselves, and humane Core Drive 8: Loss & Avoidance messaging that makes the cost of not changing visible without moralising at them. The Game Technique that does social liberation’s job here is Conformity Anchor (Social Norming, #58), a display of how people similar to the user are behaving. Remember what decisional balance says about this room: the perceived pros have to rise about twice as far as the cons fall, so the lobby is mostly about making the upside personal, not about removing friction. The Precontemplation lobby is not the Discovery phase — the user is not yet trying to choose your product — it is the stage before they are even looking. Designing it is the single highest-leverage addition TTM makes to the Octalysis toolbox.

Move 2: Match Contemplation to Discovery and Preparation to Onboarding

Once the user has acknowledged there is a problem and has started weighing change, they are in Contemplation and they are arriving at the Discovery phase of an Octalysis-mapped product. Discovery is the phase where they decide whether to engage. The cognitive-experiential Processes of Change — self-reevaluation, environmental reevaluation, dramatic relief — are exactly the right verbs for this phase. On the Core Drive side, Core Drive 1: Epic Meaning & Calling is still present, Core Drive 7: Unpredictability & Curiosity earns its keep (what happens if I do this?), and Core Drive 3: Empowerment of Creativity & Feedback appears in the form of letting the user explore the product before committing. Preparation maps to Onboarding: the user is ready to act, and the product’s job is to give them a thirty-day runway and not a sixty-second interactive tour. Stage-matched Onboarding recognises that Preparation is an active, planning stage; generic Onboarding treats it as a hurdle to clear as fast as possible. That is the first hypothesis to test, with thirty-day retention as the metric. Two canonical Game Techniques fit the runway: Step-By-Step Overlay Tutorials (#6) for the planning work, and Milestone Unlock (#19) to make the first small wins visible.

Move 3: Wire Action into Scaffolding and Maintenance into Endgame

Scaffolding is Octalysis’s name for the long middle of the product — the day-to-day use after the user has become a regular. This is where the behavioral Processes of Change earn their keep: reinforcement management, counter-conditioning, stimulus control, helping relationships, self-liberation. It is also the phase in which Core Drive 2: Development & Accomplishment takes over as the dominant Core Drive, supported by Core Drive 5: Social Influence & Relatedness for the peer-accountability dimension. Progress displays earn their keep here too, because effort tends to accelerate as people near a goal (the goal-gradient effect). Maintenance maps to Endgame, the phase Octalysis insists on designing and most products ignore. Endgame is where Core Drive 4: Ownership & Possession and Core Drive 1: Epic Meaning & Calling carry the weight, because the user is no longer extrinsically motivated to show up — they show up because the behavior has become part of who they are. The canonical Game Techniques that suit a Maintenance phase include Trophy Shelf (#64), a passive display of achievements the user is proud of; Avatar (#13), a representation of the user that accumulates ownership; Mentorship (#61), which turns a maintainer into the helping relationship for someone earlier on the staircase; and Rockstar Effect (#92), a hard-won status that makes the change part of identity. Most products ship a pretty Onboarding and no Endgame, which is why their Maintenance-stage users silently drift off.

Move 4: Design explicit spiral re-entry points

The humane piece of TTM — the spiral view of relapse — forces you to design for users who have already tried and failed at least once. Every Octalysis-mapped product should have an explicit re-entry point at each stage: a Precontemplation-to-Contemplation doorway for users who ghost, a Contemplation-to-Preparation doorway for users who stall in research mode, and a dedicated relapse doorway that drops the returning user back to Contemplation rather than Precontemplation and skips the introductory content they have already seen. On the Core Drive side, the re-entry point is a place where Core Drive 5: Social Influence & Relatedness in its helping-relationship form does more work than anywhere else in the product. “Welcome back, you’re not starting over” is one of the most valuable sentences a behavior-change product can write, and most products never write it because they are still pretending every user is a first-timer.

The canonical list already holds the parts for this doorway. Royal Return (#172) gives a user who comes back after a long absence a rewarding welcome; Return Reward Pass (#173) stretches that welcome into a short journey of rewards so the second attempt has its own early wins. Inside the habit loop itself, Streak Ramp Down (#111) lets a broken streak decay gradually instead of resetting to zero, and Revival Potion (#119) lets the user reverse the penalty once. Each one encodes the spiral in mechanics: a lapse costs something, but it does not send the user back to the lobby.

Practical Steps to Apply the Transtheoretical Model

If you are trying to take TTM from a slide deck to a shipped build this quarter, the sequence I give teams is this.

Step 1: Stage-audit your current users. Run a short readiness survey against a representative sample of your top-of-funnel traffic and your current userbase. Two or three well-designed questions will place most respondents on the five-stage staircase with acceptable accuracy. The outcome you care about is the shape of the distribution: Precontemplation / Contemplation / Preparation / Action / Maintenance. You will almost certainly find that a larger chunk of your funnel than you expected is in Precontemplation or Contemplation, and you will almost certainly find that your current product assumes most of them are in Preparation.

Step 2: Audit your current content by stage. Take every piece of onboarding copy, every email in the lifecycle series, every landing page, and every in-product prompt and label it with the stage it serves. Count. If the ratio is heavy on Preparation / Action content and thin on Precontemplation / Contemplation content — which it almost always is — you have just found the leak. Fix the leak before you build anything new.

Step 3: Pick the two highest-impact stage transitions to design for first. The two transitions that typically move the most revenue in a commercial product are Precontemplation-to-Contemplation (the awareness-to-interest step) and Action-to-Maintenance (the first ninety days of retention). Pick one of those, design the stage-matched interventions for it using the Processes of Change as your menu, and instrument the move so you can measure lift.

Step 4: Instrument self-efficacy and decisional balance. You do not need the formal instruments from the literature. Two-to-three light-touch signals inside the product — a confidence question, a motivation slider, a pros-vs-cons prompt — are enough to produce a running read on where users sit relative to the next stage transition. Use the signals to time your nudges.

Step 5: Design an explicit relapse re-entry. Work out, in advance, what the user experience looks like for a user who left, tried and failed, and is willing to come back. Drop them into Contemplation, not Precontemplation. Credit their prior progress. Name the fact that most sustained change takes several cycles. The re-entry experience is one of the most undervalued assets in a behavior-change product, and it is far cheaper to test than another acquisition campaign.

Step 6: Map each stage to its Core Drives and Experience Phase. Using the Move 1–4 map in the previous section, walk every stage of your funnel through the Core Drives it should be deploying and the Octalysis Experience Phase it belongs to. Produce a single slide that shows stage, phase, primary Core Drives, and the Game Techniques the stage calls for. That slide becomes the operating spec for every new feature you ship for the next quarter.

Step 7: Ship, measure, iterate. Stage-based design has the measurable advantage that every hypothesis you form is falsifiable in the funnel data. Set the two metrics per stage transition, ship the stage-matched intervention, and compare against the stage-generic control. You will either see the lift or you will see the stage you mis-diagnosed; either outcome is a learning you could not have gotten without the staging in the first place.

Closing Thoughts

The Transtheoretical Model has aged better than most psychological frameworks of its generation because its core claim — that the same intervention does not work for everyone, and the thing that determines what works is where the person is in the change process — is one of those claims that is obvious as soon as you hear it, and impossible to un-hear. Prochaska and DiClemente gave us a vocabulary to talk about the staircase, a menu of processes to run on each step, and the humane permission to treat the relapses as part of the climb.

Paired with the Octalysis Framework, the model stops being a diagnostic lens and starts being a build spec. The five stages are where your user is. The four Experience Phases are where your product is. The Core Drives are the levers that move one into the other. The Game Techniques are the actual widgets you ship. Do all of that, and you will have built an engagement system that addresses the full population you have instead of the self-selected few who arrived ready to change. That is the design ethic worth building toward, and it is the ethic I have been trying to encode into the Octalysis Framework from the beginning.

If you want the full version of this operating system — every Core Drive, every Game Technique, every Experience Phase applied to dozens of behavior-change case studies — the canonical reference is Actionable Gamification: Beyond Points, Badges, and Leaderboards. If you want my team to apply it to a specific product, public-health programme, or change initiative, that is what the Octalysis Group does.

Frequently Asked Questions

What is the Transtheoretical Model in simple terms?

The Transtheoretical Model, developed by James Prochaska and Carlo DiClemente in the late 1970s and early 1980s, describes behavior change as a progression through five stages — Precontemplation, Contemplation, Preparation, Action, and Maintenance — and identifies ten Processes of Change that accelerate movement between stages. The central claim is that the right intervention depends on which stage the person is in, and that stage-matched interventions should outperform generic ones, a prediction the large trials have supported only weakly.

What are the five stages of change?

The five canonical stages are Precontemplation (no intention to change in the next six months), Contemplation (intention to change within six months but not imminently), Preparation (intention to change within thirty days, with early steps underway), Action (visible behavior change sustained for less than six months), and Maintenance (behavior sustained for more than six months). Some formulations add a sixth stage called Termination, in which relapse is no longer considered a meaningful risk.

Is the Transtheoretical Model scientifically supported?

Partly. The descriptive machinery holds up best: decisional balance shifts the same way across 48 health behaviors (Hall & Rossi, 2008), and behavioral processes of change peak in Action and Maintenance across every behavior studied (Rosen, 2000). The early-stage process pattern varies by behavior, and the stage distribution varies widely by behavior and country. The intervention claim is the weak point. A Cochrane review of 41 smoking trials found stage-based versions neither more nor less effective than standard versions of the same intervention, and a meta-analysis of 33 physical-activity trials found no significant difference between stage-matched and non-matched TTM programs. The fair summary is that the model is a useful diagnostic and that tailoring to stage adds less than early enthusiasm implied.

What are the ten Processes of Change?

The ten processes are divided into cognitive-experiential and behavioral groups. The five cognitive-experiential processes are consciousness raising, dramatic relief, self-reevaluation, environmental reevaluation, and social liberation. The five behavioral processes are self-liberation, reinforcement management, helping relationships, counter-conditioning, and stimulus control. Cognitive-experiential processes dominate in the early stages; behavioral processes dominate in Action and Maintenance.

How does the Transtheoretical Model handle relapse?

Relapse is treated as a normal, expected event rather than a terminal failure. Prochaska, DiClemente and Norcross (1992) describe people as typically recycling through the stages several times before a change holds, which is why the model is often drawn as a spiral. In smoking and exercise data, only about 15% of people who relapse regress all the way to Precontemplation; most return to Contemplation or Preparation. Designing explicit re-entry points for relapsed users is one of the highest-leverage applications of the framework.

How do I apply the Transtheoretical Model to product design?

Start by staging your current users with a short readiness survey and audit your existing content by which stage it serves. You will almost always find more Precontemplation and Contemplation users than your product assumes, and a content mix heavy on Preparation and Action material. Fill the Precontemplation lobby, match the cognitive-experiential Processes of Change to the early stages and the behavioral ones to the late stages, and design an explicit relapse re-entry point. Pair the staging with the four Octalysis Experience Phases to keep the Core Drives and Game Techniques stage-appropriate.

How does the Transtheoretical Model relate to the Octalysis Framework?

The five stages map almost cleanly onto the four Experience Phases of Octalysis with one structural addition: a Precontemplation lobby that sits before Discovery and that the canonical four phases do not explicitly name. Contemplation maps to Discovery, Preparation to Onboarding, Action to Scaffolding, and Maintenance to Endgame. Each stage calls for different primary Core Drives — Epic Meaning (CD1) and Loss & Avoidance (CD8) in the early stages, Development & Accomplishment (CD2) and Social Influence (CD5) in the middle, Ownership (CD4) and Epic Meaning (CD1) in Endgame.

Is the Transtheoretical Model only for health behaviors?

TTM originated in smoking-cessation research and has been most heavily applied to health behaviors, but the framework has since been used for organisational change, education, financial-behavior change, environmental-sustainability behaviors, safety culture, and commercial product engagement. The five-stage architecture is behavior-agnostic; the thirty-day and six-month time horizons may need re-tuning for behaviors with faster or slower natural change cycles than quitting smoking.

What are the most common mistakes when applying the Transtheoretical Model?

The most common mistake is stopping at the five-stage picture and never adopting the ten Processes of Change, the self-efficacy construct, or the decisional-balance construct. The second most common mistake is treating the stages as sharp causal boundaries when the data supports them only as useful heuristics. The third is designing exclusively for Action-stage users, which leaves the majority of the target population unserved. The fourth is treating the first relapse as a failure rather than as a normal station on the spiral.

Who were James Prochaska and Carlo DiClemente?

James O. Prochaska (1942–2023) was an American clinical psychologist, Professor Emeritus of Psychology at the University of Rhode Island and a founder of its Cancer Prevention Research Center; he died in July 2023. Carlo C. DiClemente is an American clinical psychologist and Emeritus Professor of Psychology at the University of Maryland, Baltimore County. The two worked together at the University of Rhode Island from the late 1970s, published the theoretical case for a transtheoretical therapy in 1982 and the stages-of-change smoking study in 1983, and elaborated the framework over the following two decades. The 1994 book Changing for Good, co-authored with John Norcross, brought the model to a broad public audience.

References

1. Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research & Practice, 19(3), 276-288.
2. Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51(3), 390-395.
3. Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to addictive behaviors. American Psychologist, 47(9), 1102-1114.
4. Prochaska, J. O., Norcross, J. C., & DiClemente, C. C. (1994). Changing for Good: A Revolutionary Six-Stage Program for Overcoming Bad Habits and Moving Your Life Positively Forward. William Morrow.
5. Velicer, W. F., Prochaska, J. O., Fava, J. L., Norman, G. J., & Redding, C. A. (1998). Smoking cessation and stress management: Applications of the Transtheoretical Model. Homeostasis, 38, 216-233.
6. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
7. Janis, I. L., & Mann, L. (1977). Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment. Free Press.
8. Bridle, C., Riemsma, R. P., Pattenden, J., Sowden, A. J., Mather, L., Watt, I. S., & Walker, A. (2005). Systematic review of the effectiveness of health behavior interventions based on the transtheoretical model. Psychology & Health, 20(3), 283-301.
9. Cahill, K., Lancaster, T., & Green, N. (2010). Stage-based interventions for smoking cessation. Cochrane Database of Systematic Reviews, (11), CD004492. Cochrane summary
10. West, R. (2005). Time for a change: Putting the Transtheoretical (Stages of Change) Model to rest. Addiction, 100(8), 1036-1039.
11. Sutton, S. (2001). Back to the drawing board? A review of applications of the Transtheoretical Model to substance use. Addiction, 96(1), 175-186.
12. Rosenstock, I. M. (1974). Historical origins of the Health Belief Model. Health Education Monographs, 2(4), 328-335.
13. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.
14. Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4), 227-268.
15. DiClemente, C. C., Prochaska, J. O., Fairhurst, S. K., Velicer, W. F., Velasquez, M. M., & Rossi, J. S. (1991). The process of smoking cessation: An analysis of precontemplation, contemplation, and preparation stages of change. Journal of Consulting and Clinical Psychology, 59(2), 295-304. Record
16. Prochaska, J. O., DiClemente, C. C., Velicer, W. F., & Rossi, J. S. (1993). Standardized, individualized, interactive, and personalized self-help programs for smoking cessation. Health Psychology, 12(5), 399-405. Record
17. Prochaska, J. O., Velicer, W. F., Rossi, J. S., et al. (1994). Stages of change and decisional balance for 12 problem behaviors. Health Psychology, 13(1), 39-46. Record
18. Velicer, W. F., Fava, J. L., Prochaska, J. O., Abrams, D. B., Emmons, K. M., & Pierce, J. P. (1995). Distribution of smokers by stage in three representative samples. Preventive Medicine, 24(4), 401-411. Record
19. Rosen, C. S. (2000). Is the sequencing of change processes by stage consistent across health problems? A meta-analysis. Health Psychology, 19(6), 593-604.
20. Marshall, S. J., & Biddle, S. J. H. (2001). The transtheoretical model of behavior change: A meta-analysis of applications to physical activity and exercise. Annals of Behavioral Medicine, 23(4), 229-246.
21. Painter, J. E., Borba, C. P. C., Hynes, M., Mays, D., & Glanz, K. (2008). The use of theory in health behavior research from 2000 to 2005: A systematic review. Annals of Behavioral Medicine, 35(3), 358-362. Springer
22. Hall, K. L., & Rossi, J. S. (2008). Meta-analytic examination of the strong and weak principles across 48 health behaviors. Preventive Medicine, 46(3), 266-274. Record
23. Romain, A. J., Bortolon, C., Gourlan, M., Carayol, M., Decker, E., Lareyre, O., Ninot, G., Boiché, J., & Bernard, P. (2018). Matched or nonmatched interventions based on the transtheoretical model to promote physical activity: A meta-analysis of randomized controlled trials. Journal of Sport and Health Science, 7(1). PMC
24. Gilbert, S. J., Gollwitzer, P. M., Cohen, A.-L., Oettingen, G., & Burgess, P. W. (2009). Separable brain systems supporting cued versus self-initiated realization of delayed intentions. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35(4), 905-915. Record
25. Tricomi, E., Balleine, B. W., & O’Doherty, J. P. (2009). A specific role for posterior dorsolateral striatum in human habit learning. European Journal of Neuroscience, 29(11). PMC
26. Wilson, S. J., Creswell, K. G., Sayette, M. A., & Fiez, J. A. (2013). Ambivalence about smoking and cue-elicited neural activity in quitting-motivated smokers faced with an opportunity to smoke. Addictive Behaviors, 38(2), 1541-1549. Record
27. Fogg, B. J. (2019). Tiny Habits: The Small Changes That Change Everything. Houghton Mifflin Harcourt.
28. University of Rhode Island. (2023, July). University saddened by loss of Professor Emeritus James O. Prochaska. URI News
29. Cancer Prevention Research Center. Transtheoretical Model: Detailed overview. University of Rhode Island. web.uri.edu
30. Chou, Y. (2015). Actionable Gamification: Beyond Points, Badges, and Leaderboards. Octalysis Media.

Related Reading

• The Octalysis Framework: A Complete Gamification Analysis
• Self-Determination Theory: Autonomy, Competence, and Relatedness
• Dual Process Theory: System 1 / System 2 for Designers
• Maslow’s Hierarchy of Needs in Engagement Design
• Motivational Interviewing: Miller & Rollnick’s Method for Ambivalence
• Actionable Gamification: The Book

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?