When Every Bank Looks the Same

Products converge. A point of view does not.

By Roman Balzan, Chief Marketing and Brand Officer, Alpian

Last verified: September 2026

In July I wrote here about AI in banking and the question our industry still hasn’t settled: who answers when the machine gets it wrong in front of a client. That piece drew on the Digital Advantage study by Information Factory and finews.ch, which assessed 23 Swiss banks. The same study contains a quieter finding that I think matters just as much. Swiss banking offerings are becoming comparable.

Anyone who has opened a few banking apps recently knows it is true. An account, a card, an ETF portfolio, a smooth onboarding flow: a few years ago these set banks apart. Today they are table stakes. When the products look the same, a client naturally asks the question every bank should be asking itself: why choose one over another?

The instinctive answer in our industry is more features. But features get copied faster than they get launched. Differentiation has moved out of the product and into the way a bank is organised.

You can organise a bank around its product catalogue, where every client conversation is ultimately a route to something on the shelf. Or you can organise it around the client’s goal, where the question is what this person is trying to build and the products are simply the means. Alpian chose the second: banking, investing and retirement planning in one place, with one advisor who sees the whole picture. A catalogue can be copied in a year. A clear point of view on what the client is trying to achieve, held consistently over years, is far harder to copy.

The research is also honest about onboarding: across Swiss banking, it has become fast. Opening an account is no longer the hard part. The hard part is the distance between an opened account and a relationship that is actually used.

Acquisition is visible and easy to celebrate. Activation is slower and quieter. It lives in the first real question answered well, the first investment that feels understood rather than sold, the first falling market met by a person rather than a notification. This year we moved serious resources from acquisition into exactly that stretch, because a bank that wins the sign-up and loses the relationship has only rented a client. And those moments point to something larger.

If products converge, what is left to differ on? My answer: the moments where a person carries the weight. AI makes knowing cheap, so the value moves to deciding, to answering for the decision, and to knowing what to ask in the first place. Humans become more important wherever being wrong has a name attached, or wherever the answer depends on something that is not in the data. Three places make this concrete.

The exception. Models are built for the median client. Think of the widow who has inherited a portfolio she never managed, the client who is clearly being scammed but insists the transfer is fine, or the entrepreneur whose income looks like chaos on paper. Automation handles the pattern; the exception is where a bank shows what it actually is. And exceptions do not shrink as automation grows. They become the job.

What the client does not say. An advisor hears the pause before “we’re fine, really”, notices the question behind the question, remembers that a daughter starts university next year. The data holds what was recorded. A person holds the context around it.

Knowing when the machine is wrong. This becomes more valuable every time the machine improves. When output is mostly right, spotting the one wrong thing is rare, and it takes judgment built through experience. The reviewer becomes the scarcest person in the building.

The pattern underneath is that human work shifts from volume to weight: fewer hours spent producing, and more of what remains spent in moments that decide something. That is also the honest answer to the jobs debate. The work does not vanish. It concentrates, and every remaining human moment carries more. A competitor can copy your app. It cannot copy the quality of those moments.

The study also found that ecosystem integration is under-used across Swiss banking. For a young bank, partnerships are credibility extended by someone who already holds the client’s trust. When Swisscard chose Alpian as the banking partner for American Express cardholders, it lent us something much harder to build than reach.

We learned a deeper version of this when radicant stopped operating and its clients came to Alpian. Trust earned elsewhere can become trust inherited, but only if the receiving bank treats those clients as people rather than as a migrated data set. Inherited trust is honoured in the first few interactions, or it is gone.

The next generation will not discover its bank in a branch, and increasingly not on a page of search results either. It will ask an AI. “Which Swiss bank should I use to start investing?” now gets an answer, not a list of links. If a bank is not in that answer, it is not in the consideration set.

A language model describes a bank from what the world has written about it: its own publications, the press, comparison sites, reviews, years of accumulated context. You cannot buy your way into that answer. You earn it slowly, through content that is consistent, credible and genuinely useful. It is one reason we publish financial education openly in four languages, and why we now measure how the major AI models describe Alpian: what they say, which sources they draw on, where we are missing. For a bank our size, that is not a side project. It is the front door.

In a market of comparable products, the brand is the one asset a competitor cannot build in a year. We launched as Switzerland’s first digital private bank, a promise of access. Our launch campaign, No Two Alike, used real identical twins to make one point: if twins can lead completely different financial lives, nobody should get one-size-fits-all banking. It won a German Brand Award in 2023. Today our philosophy is Banking. Upgraded.: keep what Swiss banking does well, its security and its standards, and make the experience measurably better. Every product, feature and message has to pass one question: is this an upgrade? If not, we do not ship it, and we do not say it.

The biggest risk to a challenger brand is not a competitor. It is drift. Earlier this year we caught ourselves starting to sound like the incumbents we were built to challenge. So we went back to one voice, one communication bible, and clear rules for every channel and all three national languages. This is the human question again in another form. When content becomes infinite, someone has to decide what is good enough to carry the bank’s name. AI checks every piece against our standards at a speed a team our size could never manage by hand. A person still decides, and a person still signs.

Our founding idea was that wealth goes beyond money: it lies in freedom, purpose and belonging. It is the last and perhaps deepest reason humans matter more, not less. A machine optimises whatever goal you hand it. It cannot tell you whether the goal is right. Retiring at 60, buying the house, funding a sabbatical, leaving something behind: deciding what wealth is for is human work, and an optimised portfolio aimed at the wrong life is still a failure.

That is why we invest in financial education and communities such as SheWealth before asking for business, and why wealth advisory at Alpian starts at CHF 2’000, with a real advisor for every client. We are not here to bank the rich. We are here to grow the ambitious.

When products converge and content becomes infinite, trust becomes the scarce thing, and people are how trust is earned. A strong brand used to win the shelf. Now it wins the answer, and it keeps the client in the moments no machine should carry alone.

Sources: Information Factory x finews.ch, "Digital Advantage: NextGen Operating Models" (2026).