There's a moment in product work that nobody warns you about. It's not the launch that crashes or the feature nobody uses. It's quieter than that. It's the moment you realise you've been solving for the wrong version of your customer.

Mine happened while I was building an executive analytics dashboard for a digital Islamic bank in Malaysia. We were deep in the data — transaction histories, deposit patterns, account activity — trying to build a model that could categorise customers by income stability. The logic seemed sound: look at how money flows in and out, and you can infer who's financially stable, who's irregular, who's at risk.

Then I saw the pattern that broke the model.

The customer who looked dormant but wasn't

A segment of our users showed almost no regular salary credits. Occasional large inbound transfers. Low transaction volumes. Sporadic withdrawals. By every metric we had, these looked like disengaged customers — irregular at best, dormant at worst.

But something didn't sit right. These same users were depositing significant lump sums. They weren't behaving like people who had forgotten about the app. They were behaving like people who were using it very deliberately.

That's when it clicked: we weren't their main bank. We weren't even their second bank. We were their third — maybe their fourth.

This is normal. We just weren't designing for it.

In Malaysia, it's common for a person to hold three to five bank accounts simultaneously. Their salary goes into Maybank or CIMB. They might have a secondary account for specific expenses. And then there's the digital bank — the new one — which they opened because the promotional deposit rate was attractive, or the cashback was good, or they were simply curious.

This behaviour has a name in the industry: deposit arbitrage. Customers move money to wherever the returns are best, and move it back out when something better comes along. They're not disloyal. They're rational. Often, they're the most financially literate customers you have.

But here's the problem: if your analytics framework assumes you're looking at a customer's full financial life, you will misread almost everything.

That large inbound transfer? It's not income. It's a savings sweep from their primary bank. The absence of salary credits? The salary was never coming here. The low transaction volume? They don't buy groceries with this account. They park money in it.

We were reading their behaviour through a lens that assumed we were the centre of their financial world. We weren't. And that assumption was quietly distorting every insight we tried to draw.

Why this matters beyond the data

This isn't a niche Malaysian problem. Across Southeast Asia — and increasingly in markets like the UK and Europe where neobanks have proliferated — multi-account behaviour is the norm, not the exception. Customers are sophisticated. They know how to use multiple institutions to their advantage.

The uncomfortable truth for most digital banks is this: you may have acquired a customer without acquiring their financial life. Their salary, their primary spending, their savings goals — all of that may be living somewhere else. What you have is a slice. A fragment. A test.

And yet most product teams build as if they have the whole picture.

The bigger lesson

I came into fintech from architecture. In architecture, you learn very quickly that people don't move through buildings the way the floor plan says they should. They take shortcuts. They avoid the main entrance. They gather in the corridor instead of the meeting room the architect designed for them.

Finance is the same. Customers don't use banking products the way product teams assume they will. They open accounts for reasons you didn't anticipate. They hold multiple relationships simultaneously. They distribute their financial lives across institutions in ways that make sense to them, even if it makes your data messy.

The job isn't to force them into your model. The job is to understand the model they're already living in.

And sometimes that starts with a humbling realisation: you're not their main bank. You might never be. What you do with that truth is what separates products that earn loyalty from products that just chase it.

In the next piece, I'll explore what this realisation actually means for how you build — from onboarding to engagement metrics to retention strategy.