Podcast 153: Why behavioral science is the missing layer in fintech, with Anna Nyvelt & Noemi Molnar

For most of the last decade, fintech has been built around the assumption of a rational user. A user who reads the terms, compares the rates, weighs the risks, and clicks the right button.

That user does not exist.

In this episode of the Fintech Garden Podcast, Anna Nyvelt and Noemi Molnar of BeHive walk through what financial product teams keep missing, and what behavioral science actually changes when it moves from the marketing department into product and engineering.

The conversation reframes what good fintech design looks like in 2026.

90 Percent of Decisions Are Already Made Before the Click

People make around 35,000 decisions a day. Roughly 90 percent of them happen unconsciously.

That number is not a curiosity. It is a design constraint.

Every onboarding, every payment confirmation, every savings prompt is being interpreted by a brain that is mostly running on autopilot. The visible behavior — the click, the swipe, the conversion — sits on top of an iceberg of social influence, cognitive bias, emotion, and context.

Most fintech products are designed for the tip.

BeHive’s working principle is the opposite. Understand people as they are, not as they should be. That reframing changes which features get built, which flows get cut, and which metrics actually matter.

Behavioral Science Is Not the Layer You Add at the End

There is a recurring pattern in financial services. Build the functional product. Make sure it works. Then bring in marketing to figure out how to sell it.

BeHive argues that this sequence is exactly backward.

A financial product carries four layers of perceived value. Functional. Social. Emotional. Personal. Each one has to be designed in, not bolted on.

Revolut’s Ukrainian flag card. Its sponsorship of UK women’s football. The eSIM positioning that shows users a black Instagram story instead of a price comparison.

None of these is a marketing asset. They are product decisions made by teams that understood the layers.

The companies treating messaging as a finishing touch are losing to the companies treating it as architecture.

Demographics Are a Sketch. Behavior Is the Picture.

Predicting how someone will use a financial product based on age, income, or postcode is, in Anna’s framing, like predicting their decisions from their height or eye color.

Two users with identical demographics can behave in completely different ways. Why they behave that way is not in the demographic data. It is in three psychographic factors that BeHive has isolated as the most predictive in banking.

Openness to innovation. Financial risk aversion. Short-term versus long-term goal orientation.

Layered on top of the COM-B model — capability, opportunity, motivation — these factors give product teams something demographic segmentation cannot. A reason.

In environments where personalization is the differentiator, that reason is the asset.

Money Has Gone Abstract. Spending Has Gone Unconscious.

Digital money is harder to feel. There is nothing to hold, nothing to count out, nothing to physically hand over.

The result is a population that is more prone to bias, more susceptible to nudges, and less able to act in its own long-term interest than at any previous point.

Neuroscience makes this concrete. Paying activates the insula, the same brain region associated with fear. Embedded finance suppresses that signal. Buy-now-pay-later removes it almost entirely.

A user who would never take out a loan to fund dinner will accept “pay in 4” without hesitation. The product is the same. The framing is what the brain responds to.

This is the most under-regulated surface area in fintech. It will not stay that way.

Friction Is Not Always the Enemy

The default assumption in fintech UX is that fewer steps are better. Lower friction, higher conversion, cleaner journey.

The episode pushes back on this with an argument that should sit in every product review. Sometimes friction is the feature.

A two-second elevator wait pushes people toward the stairs. A one-decision pause pushes a user toward a better choice. Beehive’s research puts the customer’s tolerance for friction at roughly two decisions per flow — narrow, but real. That is the design budget for protecting users from themselves.

The line between strategic friction and dark patterns is intent. One uses behavioral science to nudge toward better decisions. The other uses it to manufacture worse ones.

Regulators arrive late to this distinction. Product teams should arrive early.

Build for Emotional Safety, Not Just Functional Completeness

Money triggers fear. That is not a metaphor.

When users open a banking app, the underlying state is anxiety. When users approve a payment, the underlying state is loss aversion. Products that ignore this build for the wrong audience.

Products that recognize it become trusted advisors.

Monzo’s savings visualizations show progress as a bike being assembled, piece by piece. The Save More Tomorrow program, designed by Richard Thaler, moved average savings from 3 percent to nearly 14 percent by allocating future raises rather than current income, sidestepping present bias entirely.

Both are small interventions. Both compounds.

That is what designing for the human mind actually produces. Not better dashboards. Different outcomes.

The Companies Winning Long-Term Are Designing for Trust

The closing argument from both Anna and Noemi is the one that tends to get pushed aside in growth-driven roadmaps.

Trust is the only metric that compounds.

Short-term optimization for conversion at the expense of the user produces a measurable long-term cost. Educated, engaged, well-served users return. Manipulated ones leave or, eventually, regulate.

The recommendation is direct. Read Thinking, Fast and Slow. Read Nudge. But work with formally trained behavioral scientists before shipping anything that touches real users at scale, because the difference between ethical nudging and dark patterns is not always visible from inside the team.

Fintech is digitizing rapidly. It is not becoming less human.

The next phase of competitive advantage will not come from faster onboarding or cleaner UI. It will come from the teams who design for the user that actually exists.