Stop Nudging Your Customers. Start Boosting Them.

The personalisation industry has spent a decade perfecting the art of the nudge. Here’s why it’s time to build something better, and what a 40-year-old psychological framework tells us about how to do it.

There’s a question that’s been bothering me for a while, and it crystallised recently during a panel discussion at London Tech Week. The question is this: when we personalise a customer experience, are we helping people make better decisions, or are we just getting better at steering them toward the decision we want them to make?

Most of the personalisation infrastructure that organisations have built over the past decade falls firmly into the second category. It’s sophisticated, data-rich, and increasingly powered by AI. But it’s also fundamentally built on a paradigm that treats the customer as a set of biases to be exploited rather than a human being to be understood.

That paradigm has a name. It’s called the nudge. And while it’s served the industry well, I think we’ve reached the point where it’s not just insufficient, it’s actively holding us back from what personalisation could actually be.

The Nudge Paradigm

The nudge, as conceptualised by Thaler and Sunstein, is elegant. Don’t restrict choices. Don’t mandate behaviour. Just design the choice architecture so that the easier, more convenient, more default option is also the better one. Opt-out instead of opt-in. Healthy food at eye level. Social proof on the checkout page.

It works. Nudge theory has improved retirement savings participation, organ donation rates, and tax compliance. In marketing, it’s become the invisible hand behind every recommendation engine, every conversion-optimised landing page, every “customers who bought this also bought…” widget.

But here’s what’s easy to miss when you’re optimising click-through rates: a nudge is, by definition, a manipulation of the environment designed to produce a specific behavioural outcome. It doesn’t change how the person thinks. It doesn’t build their capacity to make better decisions. It just makes the desired behaviour more likely in that specific context.

Change the context, and the nudge stops working. Which means you have to keep nudging. Forever. With increasingly sophisticated data and increasingly invasive tracking, because the underlying decision-making capacity of the customer hasn’t changed at all.

That’s not personalisation. That’s dependency.

Enter Boost Theory

In 2017, Ralph Hertwig and Till Grüne-Yanoff published a paper that should have changed the conversation about behavioural intervention but largely didn’t, at least not in marketing. They proposed an alternative to the nudge they called the boost.

Where the nudge changes the environment to steer behaviour, the boost changes the person’s competence to enable better decisions. Instead of making the “right” choice easier, you make the person more capable of choosing well, regardless of the context.

The distinction matters enormously. A nudge says: “We’ll design the menu so you pick the salad.” A boost says: “We’ll help you understand nutrition well enough that you’d have chosen the salad anyway.”

Hertwig and Grüne-Yanoff identified three types of boost:

Cognitive tools — teaching people heuristics, mental models, and decision frameworks that transfer across contexts. Not “here’s what to buy,” but “here’s how to evaluate what’s worth buying.”

Representational boosts — changing how information is presented so that people can process it more effectively. Natural frequencies instead of percentages. Visual risk representations instead of abstract probabilities. Making the complex comprehensible without dumbing it down.

Competence boosts — building transferable decision-making skills. Statistical literacy. Risk assessment. The ability to evaluate evidence. Not training for a specific decision, but building the capacity for better decision-making in general.

The boost doesn’t treat the customer as a bias to be exploited. It treats them as a person to be equipped. And that distinction changes everything about how you think about personalisation.

Why This Matters Now

The nudge paradigm has dominated marketing for a decade because it’s measurable, optimisable, and it works in the short term. You can A/B test a nudge. You can measure the conversion lift. You can put it on a dashboard and show the board.

But three things have changed that expose its limitations.

First, customers are becoming aware of being nudged. The more sophisticated the manipulation, the more it feels like manipulation when it’s noticed. Trust erodes. Ad blindness was the first signal. The growing backlash against hyper-personalisation is the second.

Second, AI is about to change who makes the decision. If an AI agent is acting on behalf of the customer, evaluating options within pre-approved parameters, traditional nudges are largely irrelevant. The agent doesn’t respond to social proof, emotional framing, or default options. The human who set the parameters might, but that’s a different moment in the decision journey, and one that most personalisation infrastructure isn’t designed for.

Third, and this is the one that keeps me up at night, we’ve built an entire industry optimisation stack around making people more predictable rather than more capable. That’s fine when it works. But when the context changes, and contexts always change, predictable people make the same decisions in a world that requires different ones. We haven’t just failed to build their decision-making capacity. We’ve actively discouraged them from developing it by making all the decisions for them.

The boost offers a different path. And when you map it onto the way people actually make decisions, something interesting happens.

The STAR Framework: A Motivational Map for Boosting

This is where I want to introduce a framework I’ve been working with for over a decade, because it solves the biggest problem with boost theory: the assumption that everyone needs the same kind of boost.

The STAR Operating System is built on seven foundational psychological theories, but its core insight comes from Self-Determination Theory, which identifies three universal psychological needs: autonomy, competence, and relatedness. STAR adds a fourth, security, and maps these needs onto four primary motivational types:

The Socialiser (S) is driven by relatedness. They make decisions through connection, community, and belonging. They want to know what people like them experienced, what the group thinks, and whether the choice strengthens their relationships.

The Thinker (T) is driven by competence. They make decisions through analysis, comparison, and mastery. They want data, frameworks, and the confidence that they’ve evaluated every option systematically.

The Adventurer (A) is driven by autonomy. They make decisions through exploration, instinct, and momentum. They want options, flexibility, and the freedom to choose their own path without being boxed in.

The Realist (R) is driven by security. They make decisions through caution, provenance, and risk assessment. They want guarantees, track records, and the reassurance that the safe choice is also the smart one.

These aren’t personality types in the old psychographic sense. They’re motivational orientations rooted in fundamental human needs, the same needs that have been documented across decades of psychological research. And they determine not just what people choose, but how they choose, what information they seek, what risks they tolerate, and what “a good decision” actually means to them.

Each type branches into three archetypes, giving twelve distinct profiles. The Thinker who leads with competence but leans toward social connection becomes the Structured Collaborator. The Adventurer who leads with autonomy but tempers it with analytical rigour becomes the Inventive Pathfinder. The Realist who leads with security but channels it through social bonds becomes the Steady Harmoniser. Each archetype has its own decision-making signature, its own communication preferences, and its own relationship with risk.

This is the motivational map that boost theory is missing. Because if you don’t know what drives someone’s decisions, you can’t know what kind of competence to build.

From Type to Boost: What Each Profile Actually Needs

Here’s where the framework stops being academic and starts being practical.

Boosting the Socialiser

The Socialiser’s decision-making is socially embedded. They evaluate options through the lens of community experience, relational impact, and group consensus. But this isn’t the same as being susceptible to social proof nudges. A nudge says “87% of customers recommend this.” A boost says “here’s what people in your situation actually experienced, in enough depth for you to evaluate whether their experience is relevant to yours.”

The competence you’re building is relational intelligence. Help them distinguish between genuine community signals and manufactured consensus. Give them tools to evaluate whether the “people like them” are actually like them, or just superficially similar. Equip their social decision-making instinct rather than exploiting it.

In student accommodation, this means moving beyond star ratings and testimonials. It means giving Socialiser-motivated students access to real community stories, the messy, honest, detailed kind that lets them evaluate whether they’d actually belong there. Not curated marketing content. Real human experience, presented with enough depth to be useful.

Boosting the Thinker

The Thinker is the natural fit for boost theory because they already seek mastery. Their primary need is competence, and they’re naturally drawn to frameworks, data, and analytical tools. They don’t need to be nudged toward the “right” answer. They need to be equipped to find it themselves.

The competence you’re building is analytical rigour. Decision frameworks. Comparison tools. Data literacy. The ability to evaluate evidence, spot misleading metrics, and distinguish between genuine value and marketing noise.

In practice, this means giving Thinker-motivated students systematic comparison resources. Not “here’s our best room” but “here’s how to evaluate room options across fifteen dimensions that actually matter.” Give them the framework and they’ll make the decision themselves, better than any funnel could steer them. They’ll also trust the brand more because you respected their intelligence rather than trying to circumvent it.

Boosting the Adventurer

The Adventurer’s decision-making is instinct-driven and momentum-oriented. They evaluate options quickly, trust their gut, and want the freedom to explore without being constrained. Traditional personalisation hates Adventurers because they don’t follow predictable paths. They bounce, they explore, they make decisions that look irrational from a behavioural data perspective but make perfect sense from a motivational one.

The competence you’re building is navigational intelligence. Don’t narrow their choices (that’s a nudge, and Adventurers will actively resist it). Instead, expand their ability to evaluate choices quickly and effectively. Scenario planning. “Here’s what you’re not seeing.” Option comparison at speed. Tools that respect their pace rather than trying to slow them down.

In student accommodation, this means giving Adventurer-motivated students immersive, self-directed exploration tools. Virtual tours they can control. Neighbourhood guides that highlight the unexpected. Comparison tools that work at speed rather than requiring fifteen form fields. Let them explore freely, but make sure the exploration is rich enough that their instinct-driven decisions are well-informed ones.

Boosting the Realist

The Realist’s decision-making is risk-calibrated and provenance-driven. They evaluate options through the lens of safety, track record, and downside protection. They’re not risk-averse in the sense of being timid; they’re risk-aware in the sense of being strategic. They want to know what could go wrong before they commit.

The competence you’re building is risk assessment. Give them frameworks for evaluating what’s actually dangerous versus what just feels dangerous. Help them distinguish between genuine signals and noise. Provide contingency planning tools. Equip them to feel confident in their evaluation, not just comfortable with the outcome.

In practice, this means giving Realist-motivated students honest risk information. Not “everything is great” marketing copy, but “here’s what to look for, here’s what the data says, here’s what could go wrong, and here’s how to protect against it.” The Realist who feels genuinely informed is a confident decision-maker. The Realist who feels nudged is an anxious one who’ll either churn or never convert at all.

The Conversion Lift That Lasts

Here’s the counterintuitive part. Boosting should, in theory, be worse for short-term conversion than nudging. If you’re helping people make genuinely better decisions, some of those decisions will be not to buy. A nudge would have converted them. A boost lets them choose freely, and sometimes freely means no.

So why would any marketing team choose boost over nudge?

Because the customers who do convert after being boosted convert for the right reasons. They’re not buying because the default was set in your favour, or because the countdown timer created false urgency, or because the social proof widget made them feel like they were missing out. They’re buying because they genuinely evaluated the offer and decided it was right for them.

Those customers don’t return the product. They don’t churn after the first billing cycle. They don’t feel buyer’s remorse. And they tell other people about it, not because you incentivised the referral, but because they actually recommend the thing they bought.

The conversion lift from boosting is slower, harder to measure, and completely invisible to a last-click attribution model. But it’s real, it compounds, and it builds something that nudge-based personalisation never will: trust.

Trust is the ultimate conversion asset. And you can’t nudge your way to trust. You have to earn it by making people more capable, not more compliant.

The AI Agent Problem

This is where the boost becomes not just preferable but essential.

If the customer on the other end is an AI agent acting within pre-approved parameters, the entire nudge infrastructure collapses. The agent doesn’t respond to emotional framing. It doesn’t care about the colour of the call-to-action button. It’s not influenced by social proof, scarcity signals, or the fact that you’ve put the premium option in the middle column.

The agent evaluates against parameters: price, features, ratings, availability, compliance with the human’s stated preferences. And those parameters were set by a human, in a specific moment, with a specific set of motivations.

The question becomes: how competent was that human when they set the parameters?

If they were Realist-motivated and you helped them understand what “risk” actually means in your category, they set better parameters. If they were Adventurer-motivated and you gave them the tools to evaluate optionality, they set better parameters. If they were Socialiser-motivated and you helped them understand the community experience, they set better parameters. If they were Thinker-motivated and you gave them the analytical framework, they set better parameters.

The boost doesn’t disappear in an AI-mediated world. It moves upstream. Instead of empowering the person at the point of purchase, you’re empowering the person at the point of parameter-setting. And that moment is earlier, less visible, and more consequential than anything that happens on your checkout page.

Marketing moves from persuasion at point of conversion to empowerment at point of discovery. And the organisations that invested in building customer competence, not just customer predictability, will have a structural advantage.

What This Looks Like in Practice

Let me make this concrete.

In student accommodation, we work with young people making one of the biggest decisions of their lives, often their first time living away from home. The traditional personalisation approach would be: track their browsing behaviour, identify their price sensitivity, serve them targeted ads based on their demographic profile, and optimise the booking funnel for conversion.

That’s a nudge stack. It works in the short term. It also produces students who booked for the wrong reasons, churn at alarming rates, and leave negative reviews that damage the brand for years.

A boost approach, informed by STAR, asks fundamentally different questions. How do we help Realist-motivated students understand what “safe” actually means in a new city, beyond the marketing brochure? How do we help Adventurer-motivated students evaluate their options without overwhelming them? How do we help Socialiser-motivated students understand what the community is actually like, not just what the photos show? How do we help Thinker-motivated students compare options systematically rather than getting lost in tabs?

The answers to those questions aren’t ads. They’re tools. Decision frameworks. Honest comparison resources. Community stories with enough depth to be useful. Risk assessment guides that respect the student’s intelligence.

Some of those students will decide not to book with us. That’s fine. The ones who do book will be booking for reasons that hold up, and they’ll stay, and they’ll recommend us because they genuinely had a good experience, not because we optimised a funnel.

The Uncomfortable Truth

The hardest part of moving from nudge to boost isn’t the technology. It isn’t the data. It isn’t even the expertise.

It’s admitting that the personalisation infrastructure you’ve built is optimised for the wrong outcome.

Most personalisation stacks are designed to maximise conversion. Boost is designed to maximise customer competence. Those two objectives are not always aligned. Sometimes the most competent decision is to not buy. Sometimes the most nudged decision is to buy something you don’t need.

If your board is measuring success by conversion rate, boost looks like a step backwards. If your board is measuring success by customer lifetime value, retention, and trust, boost is the only approach that compounds.

That’s the conversation that needs to happen. Not “should we use behavioural science?” but “what kind of behavioural intervention are we building, and who does it actually serve?”

The nudge serves the organisation. It produces predictable customers who convert at higher rates in the short term.

The boost serves the customer. It produces capable customers who make better decisions and stay longer.

The organisations that figure out how to do both, how to empower the customer while still building a sustainable business, will be the ones that win. Not because they have better data or smarter algorithms. But because they asked a better question.

Instead of “how do we get them to convert?” they asked “how do we help them choose well?”

That shift changes everything. And it starts with the unlearning.

David Chadderton spent his twenties and thirties teaching people how to make life-or-death decisions at forty thousand feet. He now applies the same principles to consumer psychology, which, depending on the brief, can feel equally high-stakes. He’s the creator of the STAR Framework and the author of The STAR Framework: Rewriting the Rules of Consumer Engagement (NYC Big Book Award 2025), The STAR Operating System: Decode Mindset, Understand Motivation, Transform Human Behaviour, and Dear Algorithm, It’s Not Me, It’s You. By day, a Chief Marketing Officer. By night, a behavioural science obsessive who writes The Unoptimised Human because he can’t stop thinking about why people do what they do.