One bank, two regulatory regimes
The same institution, operating under two regulators, makes two incompatible choices about who its interface is for. Neither is wrong. Neither is exportable.
Independence. This analysis was not commissioned by, or conducted for, HSBC. It reads two publicly accessible retail websites as they stood in July 2026, using a stated method, and draws conclusions about regulatory environments rather than about the institution. No confidential, internal or client material was used.
The question
A large retail bank operating in two jurisdictions has to satisfy two regulators with different theories of what a customer needs to be told. That pressure has to land somewhere visible. The question was whether it lands in the navigation — the first structural decision any retail site makes, and the one that reveals who the interface is actually built for.
What I did
Reconstructed the full top-level navigation of both public sites, counted the navigable links by category, and mapped the equivalent customer journeys side by side. No access, no interviews, no internal data — only what any customer in either market can see, recorded on a stated date so the finding remains checkable as the sites change.
The finding
One bank operates six top-level categories in each market. Four are the same. The two each market chooses freely are not — and that choice is the whole result.
Density buys competence and pays in trust. The United Kingdom spends its discretionary space on people in difficulty; Hong Kong spends it on people with capital. Neither is wrong. Both are unexportable.
Every whole system buys something and pays for it. Naming the payment is the analysis; praising the result is not.
Why it matters commercially
This is the argument a transformation programme has to be able to make and usually cannot: that a design difference between two markets is a regulatory artefact rather than a failure of consistency, and that harmonising them would destroy something each market needs. It is reached from public evidence, with the method stated before the conclusion, so a sceptical reader can reproduce it.
What it demonstrates
Reading a system for what it trades away rather than for how it looks — and being able to show the working to someone who does not have to believe you.