Smart Friction: Why the Best Financial Products Know When to Slow Down

Over the past decade, FinTech has chased frictionlessness. Tap and pay, one-click onboarding, instant transactions. The idea was simple: removing friction would always improve the product. For younger, digitally confident customers building their financial lives, this approach largely worked.

But removing friction is not always neutral. Something else gets removed with it: the feedback loop that helps people understand what they are doing. When someone pauses before a large transfer, sees a warning, or has to confirm a decision, they start to build a mental model. They learn what is usual, what might be risky, and when to think twice. Strip that out and you do not just make the product faster. You make the user more exposed.

The financial services industry has started to recognise this. The question now is not whether to add friction back, but where, how, and for whom.

The science behind the pause

There is a reason frictionless design is so effective, and why it also carries risk.

Research consistently shows that a large proportion of purchase decisions happen on impulse. For most people, that is not recklessness. It is biology. The brain's reward system fires before rational thought catches up. Dopamine responds to speed, immediacy, and ease. That is why one-click purchasing works so well: the impulse arrives and the barrier is gone before the rational mind gets a look in.

For low-stakes, reversible decisions, that is mostly fine. You can return the shoes. You can cancel the subscription.

But some decisions do not work that way. Richard Thaler, who won the Nobel Prize for his work on behavioural economics and choice architecture, made a useful distinction here. Most nudges work in both directions. You can be nudged towards a decision and nudged back out of it. But for some decisions, the architecture of the moment matters far more, because once you have gone through, you cannot turn back. Thaler called these one-way streets.

Choice architecture, the idea that how a decision is presented shapes the decision itself, is now well established in financial services. The order of options, the default settings, the presence or absence of a confirmation step: all of these shape outcomes in ways that are measurable and significant. Regulators have picked this up. The FCA's Consumer Duty framework is partly built on the same insight: that product design is not neutral and firms are responsible for the decisions their design encourages.

Pension decisions are one-way streets. So are large transfers made under pressure. So is the decision to start drawing down income in a way that locks in your tax

position for the year. The stakes are categorically different, and a well-designed product should recognise that.

Who pays the price

Financial inclusion has typically focused on access: who can open a bank account, who can get credit, who is left out entirely. People near or in retirement rarely feature in that conversation. By most measures, they are already included. They have accounts and pensions.

But being inside the system does not mean it works for you. If we only measure access, we miss the people who have accounts but struggle with products that were never designed with them in mind.

The FCA's Financial Lives 2024 survey found that 49% of UK adults have at least one characteristic of vulnerability. The FCA defines this through four drivers: health, financial resilience, capability, and life events. Those four things cluster heavily around retirement, and the transition into retirement tends to trigger several at once. A person retiring might be dealing with a health change, a significant drop in income, a bereavement, and a set of financial decisions they have never faced before, all at the same time.

The Alzheimer's Society puts the number of people in the UK living with dementia at around 982,000, rising to 1.4 million by 2040. Cognitive change affects financial decision-making well before any formal diagnosis. A multi-step digital journey that feels straightforward at 45 can become genuinely difficult at 73. The product does not adapt. The person has to.

When bank branches closed, we lost a layer of informal protection. Which? found that older and disabled people were hit hardest. The teller who would ask 'are you sure about that?' is gone. The digital tools that replaced branches do not provide that kind of feedback. They were not designed to.

This pattern is not exclusive to retirement. It shows up whenever decisions are rare, stressful, or high-stakes: after bereavement, during income shocks, in periods of financial worry. Retirement makes it more visible, because the risks are larger, the decisions are less reversible, and the products are least equipped to help.

What smart friction looks like in practice

The industry does have examples of getting this right. Just not yet in pensions.

Monzo built a gambling block. More than 700,000 customers have turned it on voluntarily, not because they were required to, but because they wanted a pause between impulse and action. In 2024 alone, the feature blocked over £9 million in gambling transactions. The most telling detail is not the scale. It is that people asked for it. They wanted the friction. The product just had to make it available. Users could also set their own cooling-off period before turning the block off, with 75% choosing periods of more than 48 hours. The friction was not imposed. It was chosen.

Charlie, a US fintech built specifically for people over 62, adds a six-hour hold on transfers to new payees above $100, with a fraud alert during the window. Older

adults using the feature reported feeling more in control, not more restricted. The pause is short. The protection is real. And because it is calibrated to context, a new payee, a material amount, a demographic known to be targeted by fraud, it does not interfere with normal day-to-day use. Charlie was not designed around compliance. It was designed around what this specific group of users actually needed.

Both products share a design principle: the friction is smart because it fits the moment. It is not a generic warning or a blanket delay. It responds to specific signals that suggest this particular transaction, for this particular user, warrants a second look. That is the difference between friction that helps and friction that just annoys.

Generic banking infrastructure cannot do that. Confirmation of Payee checks whether a name matches an account. Reimbursement rules push banks to compensate fraud victims after the fact. These are good steps. But they do not know whether this person almost never moves large sums, or whether this is the first time she has handled money alone after years with a partner.

A frictionless retirement

Most digital products are built around habits. Drawdown is different.

Take out a lump sum in the wrong tax year and you are stuck with it. Make a rushed transfer and the money could be gone for good. Consolidate pension pots at the wrong moment and you could lose valuable guarantees you did not know you had. It is worth knowing where you are heading before you commit.

Now add the emotional context. Someone making drawdown choices might be grieving, or managing money alone for the first time after years with a partner. They might be navigating a complicated pension without any professional help. Around 750,000 people retire in the UK every year. Most of them do not have a financial adviser. They are making some of the most consequential financial decisions of their lives with no professional guidance and products that were not built for this moment.

The product has to provide the feedback loop, because many of these customers cannot build it themselves from scratch. And unlike most financial decisions, there is limited room to course-correct afterwards.

Most pension providers communicate with their members once a year, by letter. That is the sum total of the relationship at the point it matters most. When someone finally engages with their pension in decumulation, they are often doing it alone, for the first time, under time pressure, and the product they encounter was built for accumulation, not for this.

The data problem

Here is where the opportunity sits, and where the gap is most obvious.

A pension provider knows your age, the size of your pot, your contribution history, and your drawdown elections. That is rich, retirement-specific context. But it does not see your day-to-day financial behaviour. It might process one or two drawdown payments a month. That is not enough transaction history to understand what normal looks like for you or to detect when something is off.

A bank sees your spending patterns, your income flows, and what normal looks like day to day. But it knows nothing about your retirement position, how your pot is tracking, or whether a particular withdrawal will create a tax problem down the line.

Today those two data sets sit in completely separate systems. Neither can see the full picture. A pension provider cannot tell whether a large transfer is out of character. A bank cannot tell whether it puts the member's retirement at risk.

Connecting those two worlds is exactly what Freedom is building. By bridging the pension and the payments layer, we make contextual intelligence possible: a product that can draw on both what is normal for this person and what matters for their retirement. That is the foundation smart friction actually needs to work. Without it, any pause or nudge is just a generic speed bump. With it, the intervention can be shaped by what the product already knows, and calibrated to the actual decision being made.

Getting the design right

Contextual intelligence is only useful if the product is usable in the first place. The two things have to work together.

Simpler journeys, not simpler products. Decluttering an interface reduces cognitive load so the person can focus on the decision itself rather than navigating the system. Calm screens, simplified navigation, complex processes broken into smaller steps. These are not accessibility add-ons. This is good design for high-stakes decisions, and it benefits every user regardless of age or ability.

The specific design of a pause matters too. A confirmation step that asks 'are you sure?' teaches nothing. A confirmation step that says 'this transfer is larger than anything you have sent before, and it cannot be reversed once processed' gives the user something to act on. The difference is whether the friction is informative or just obstructive.

Authentication needs to be usable. SMS codes are hard for many people, including those with low vision, limited movement, or older devices. Biometrics are often easier and more reliable. Designing authentication that works for the least able user in the customer base tends to produce a better experience for everyone.

Language matters more than the industry usually acknowledges. Pension choices involve tax wrappers, nomination forms, and income options that providers have explained so poorly that regulators now require testing of customer communications. One plain screen with a single clear question is more useful than a dashboard full of numbers nobody asked for. The goal is not to simplify the product. It is to reduce the effort required to understand the decision.

When you design with users rather than for them, the features that tend to emerge are the ones people actually want: spending locks, trusted contact alerts, the ability to pause their own account, clear notifications that explain why something has been flagged. These are not niche features for vulnerable customers. They are features that build trust and confidence across the board.

This is not optional

Since 2023, Consumer Duty requires firms to demonstrate good outcomes for customers with vulnerabilities. That is not a compliance exercise. It is a design brief. The FCA has been clear that it expects firms to identify the needs of vulnerable customers at the product design stage, not as an afterthought. Firms cannot keep deferring it.

The fraud case makes the urgency equally clear. The UK lost £1.17 billion to fraud in 2024. Authorised push payment scams, where someone is pressured into approving a payment themselves, fell only after the industry added friction. The PSR's mandatory reimbursement rules are now in force, which means the cost of getting this wrong falls directly on firms. The data is unambiguous. Pauses work. The industry already knows this. The question is whether it will apply the lesson in the places where it matters most.

There is also a retention argument that the industry tends to underestimate. UK private pensions hold around £3 trillion in assets. £380 billion leaves pension platforms every year, not because members decided to leave, but because nobody built a reason to stay. If members are harmed, confused, or simply not well served by their pension product at the point they need it most, they leave. In a market where accumulation products are broadly similar, the decumulation experience is increasingly where the real competitive differentiation sits.

Pension providers were not set up to offer a banking-style experience. Fintechs aimed at younger customers have not shifted their focus either. Closing this gap properly means building for this group from the ground up, not retrofitting something designed for a different customer in a different life stage.

The design principle behind all of it

Smart friction is not a bolt-on for vulnerable customers. It is a design principle that applies wherever the stakes are high enough to warrant it.

Some decisions need space. Some journeys need to breathe. Most customers are better served by a product that slows down in the right places than one that never stops at all. The retirement context makes this obvious because the money is significant, the decisions are complex, and the margin for error is small. But the underlying logic holds across financial services wherever the cost of a rushed decision outweighs the benefit of speed.

The frictionless era was not a mistake. It produced real improvements for real customers. But it also produced a set of products that work well when everything goes smoothly and struggle when it does not. Designing for the moments when things get difficult, when the user is under pressure, when the stakes are high enough that getting it wrong actually matters, is the next design challenge the industry needs to take seriously.

The question is not "can we remove friction?" It is "will we add the right kind back?"

‍

By Prerna Goel, COO, Freedom

‍