ux metrics for wealth products: measuring adoption, engagement and task success
measure what matters
adoption, engagement, task success
what gets measured gets ... misunderstood
In finance products, metrics have a special talent for becoming vanity theatre. Someone screenshots a dashboard, circles a number and suddenly the product is “doing well”.
Meanwhile clients are still getting stuck on basic tasks like finding a statement, understanding performance, or completing a form without rage clicking.
So yes, measure your wealth or financial experience. Just don’t confuse activity with value. A spike in sessions can mean interest, or it can mean people are lost and reloading pages like it’s 2009.
three buckets that actually matter
If you want metrics that are useful, keep them in three buckets. Anything else is garnish.
adoption
Adoption is: did people use the thing at all?
Good adoption metrics:
active users or tracked instances over time
feature adoption rate (feature users divided by total users)
repeat usage within 30 days
Common trap: counting “views” as adoption. A view might be a misclick, or a dead end. Try tying adoption to a meaningful action.
engagement
Engagement is: did the experience hold attention in a healthy way?
Useful engagement signals:
engaged sessions rate
average engaged time per session
pages per session, used carefully
return frequency by segment, not just overall
Fairly obvious note: if your “engagement win” is people spending longer on a confusing form, you have not won anything. You have built a time sink.
task success
Task success is: did people complete what they came to do, without drama?
This is the bucket most teams skip because it requires thinking. And measurements we don’t always have access to.
Good task success metrics:
completion rate for core journeys (start to finish)
time to complete a task, tracked as a distribution not a single average
drop-off step in multi-step forms
error rate and recovery rate
support contact rate per journey
If you can only track one thing, track completion. A portal that does not help people finish is just a fancy brochure.
pick a small set of “jobs”
Wealth platforms can do a lot. That is the problem.
Pick a short list of core jobs and measure them properly:
log in successfully
find and download a document
check portfolio value and performance
send a secure message
complete onboarding or a form
update key details through the safe route
Then create one metric per job that tells you if it worked.
Example: “documents task success” could be:
Downloads completed divided by document detail views
It is not perfect but it is directionally honest.
don’t average your way into lies
Finance behaviour is lumpy. A small set of clients do a lot, and many do a little.
So avoid single averages as your headline. Use:
medians
percentiles (p50, p90)
segments (new vs returning, desktop vs mobile, engaged vs not)
If p90 time to download a statement is 4 minutes, that is your reality, not the average.
connect metrics to decisions
A metric without a decision attached is a hobby.
Try this simple framing:
If adoption drops, we improve findability and onboarding cues
If engagement drops, we check performance and clarity, not add content
If task success drops, we fix friction at the step level
If engagement rises, we think about a new useful feature
And always pair behavioural metrics with qualitative signals, even if it is lightweight. Support tickets, message themes and top search queries are free research. Use them.
the small dashboard i actually trust
If you want a minimal executive view, this set is practical:
tracked instances
sessions
engagement rate
average engaged time per session
documents adoption
secure messaging adoption
form completion rate
task success for one flagship journey
That dashboard tells you reach, behaviour and whether the product helps people finish. Everything else is a nice to have.
