information architecture for wealth platforms: navigation that matches mental models
invisible information architecture
match what clients want not what the business wants
everyone loves “simple nav” until it has to work
Information architecture in a wealth portal is one of those things people only notice when it is bad. When it is good, it feels obvious. When it is bad, users start doing that quiet rage thing where they click three places, pause, unconsciously blaming themselves.
They should not blame themselves. It is not their job to learn your business processes or requirements.
The reality is blunt: wealth and financial platforms are complex. Clients have accounts, portfolios, documents, messages, tasks, personal details and life stuff that does not fit neatly into a menu. Your job is not to hide complexity. Your job is to organise it so it matches how people think.
mental models beat internal models
Most navigation structures are based on what the business calls things. Clients do not care.
Clients think in tasks and questions:
how much do i have
how is it doing
what do i own
what has changed
what do i need to do
where are my documents
how do i talk to someone
If your IA does not map to those questions, you will get misclicks, drop-offs and support calls. Then someone will say “we need better onboarding” which is the usual lie we tell instead of fixing the structure.
start with the top jobs, not the feature list
A practical way to design IA for finance systems is to define a small set of “top jobs” and make them the spine.
Typical top jobs in a wealth platform:
view portfolio summary
understand performance
find and download documents
message an adviser or support
review net worth, assets and liabilities
complete tasks, forms, onboarding steps
manage profile and security
Notice what is not there: “insights”, “market news” and “learn”. Those can exist, but they should not displace the core jobs.
the menu is not the product, but it decides the product
Navigation is an attention budget. Every label you add competes for a slot in the user’s head.
Practical rules:
keep primary navigation to 5 to 7 items
use nouns for destinations and verbs for actions
avoid duplicate concepts across labels
use one label for one idea, not one label for three ideas
If you have “accounts”, “portfolios” and “investments” as separate items, you are probably making clients translate your taxonomy. Clients do not know the difference, and they should not need to.
structure patterns that work
1) layered navigation: broad then deep
Use a small set of top-level destinations, then allow users to drill down within a section. This reduces scanning overhead and stops the nav becoming a junk drawer.
Example:
portfolios
overview
performance
assets
allocations
transactions
This works because the user stays in the same mental space. They do not feel bounced around.
2) consistent location signals
Users need to know where they are, what is inside this section, and how to get back.
Good location cues:
clear page titles that match navigation labels
breadcrumbs only when hierarchy is real
left-hand sub-nav for depth, not for random links
If the page title says “account statements” but the navigation says “documents”, users will feel lost even if they are in the right place. Label consistency is a trust signal.
3) cross-links based on tasks, not marketing
Cross-links should help people complete a job. Not encourage “discovery”.
Helpful cross-links:
from a portfolio view to related documents
from a document to the message thread about it
from a net worth summary to the underlying categories
Unhelpful cross-links:
“recommended articles”
“top insights”
“explore more”
Those belong in a secondary area. Not in the path of real work.
the common pitfalls
dumping everything into “more”
If “more” becomes your fastest-growing navigation item, you have an IA problem. “More” is not a category. It is an apology.
mixing audiences in the same IA
Clients and colleagues think differently. If you are building a combined portal, do not assume one navigation works for both. If you must share IA, use role-based landing and wording that fits the user’s job.
hiding support
Support is not a footer link. In finance systems, help should be visible in the primary interface. Users should never wonder if anyone is there.
making documents a second-class citizen
Documents are a top job. Treat them like one. If users need statements for a mortgage, they will not admire your dashboard. They will judge whether they can find the file in 20 seconds.
a practical IA checklist
Can a user find their top three tasks within 10 seconds?
Do navigation labels match the words users say out loud?
Is there a single obvious home for documents?
Are portfolio-related views grouped together, not scattered?
Is support visible on every page without hunting?
Can users recover from getting lost with a clear “back to overview” path?
If you can tick those off, your IA is doing its job.
parting thoughts
A wealth portal can have beautiful charts and still fail because the navigation makes people feel stupid.
Good IA does not impress people. It disappears. And that is exactly what you want.