Once you accept that you're probably not your customer's main bank, a question follows immediately: so what do you build instead?

Most teams don't ask it. The assumption baked into roadmaps, OKRs, and investor decks is that the goal is always to become the primary account — the one where salary lands, where daily spending happens, where the full financial picture lives. That's the north star.

It's also, for many users, an unrealistic one. And designing toward it without acknowledging the reality of how customers actually use your product is how you end up building features nobody needs, measuring metrics that don't reflect value, and writing off your most financially sophisticated users as "inactive."

Daring to build for exactly who you're serving means starting with an honest reckoning with the relationship you actually have.

Rethink onboarding

If a customer is opening their third or fourth bank account, they do not need to be educated about digital banking. They've been through onboarding flows before. They know what a transaction notification looks like. They don't need a tutorial on how to transfer money.

What they need — and what most digital bank onboarding fails to provide — is a fast, clear answer to one question: why should I use this one?

The onboarding flow that works for a first-time banking customer is the wrong flow for a deposit arbitrageur. The former needs reassurance and education. The latter needs to understand your competitive edge in thirty seconds and get out of the way.

Designing one onboarding for both is designing for neither.

Fix your engagement metrics

Daily active users. Monthly active users. Session frequency. These are reasonable metrics for products where daily engagement is the intended use case — a social app, a productivity tool, a news reader.

They are the wrong metrics for a product that a financially strategic customer uses once a month to check their deposit return.

That customer — logging in on the 1st of every month, confirming their interest has accrued, logging back out — is not disengaged. They are using your product precisely as intended. Flagging them as inactive and triggering a re-engagement campaign is not just wasteful. It signals to them that you don't understand why they're here.

Engagement metrics need to reflect the actual relationship. For a savings-parking customer, the meaningful signal isn't session frequency. It's deposit retention. It's whether they moved money out — and if so, where it went and why.

Build segmentation on behaviour, not assumptions

Income stability models built on partial transaction data will systematically misclassify multi-account users.

A customer with no salary credits in your system isn't necessarily unemployed or financially unstable. Their salary is landing in another bank. A customer with irregular deposit patterns isn't necessarily unpredictable — they might be deliberately moving money between institutions based on rate changes.

Treating these users as high-risk because your data is incomplete isn't just inaccurate. It shapes how your product treats them — what features they see, what offers they receive, what limits get applied. The misclassification has real consequences.

Honest segmentation in a multi-account world requires acknowledging what you can't see. It means building models that are calibrated to partial data, and labelling insights with appropriate confidence levels rather than presenting them as complete truths.

Redesign retention for the relationship you have

You cannot retain a rate-chasing customer with a loyalty programme designed for primary account holders. Points, badges, tiered rewards — these are tools for deepening emotional investment in a product someone already lives in. They don't work on someone who is with you precisely because the return is better than it is elsewhere.

For the deposit arbitrageur, retention looks different. It's about staying competitive on the thing they came for. It's about removing friction from the experience of moving money in and out — because paradoxically, making it easier to leave is sometimes what makes customers comfortable staying. It's about being honest about what you are.

The mistake isn't having customers who use your product transactionally. The mistake is designing a retention strategy that alienates them because it was built for someone else.

Design for the relationship you have — not the one you want

The most honest version of this insight is also the most uncomfortable one for product teams to sit with: not every customer will become a primary account holder. Some of them are here for one specific reason, and when that reason disappears, they'll leave.

That's not a failure. That's a customer type. And it deserves its own product thinking — its own onboarding, its own metrics, its own retention logic, its own feature set.

The teams that will build lasting products in digital banking are the ones willing to look at their actual user base, understand the real relationships those users have with their product, and build honestly toward that — rather than toward a version of the customer that only exists in the pitch deck.

Dare to build for exactly who you're serving. It's harder than building for who you wish was there. It's also the only thing that works.