Complex systems
Cross-systems scaling
In large organizations, users rarely interact with one system at a time. They interact with aggregations of systems like alerts, dashboards and internal tools where information has to be accurate, current and understandable at a glance.
This pattern focuses on designing a single, coherent experience over several systems without hiding important detail or oversimplifying the underlying logic.
Wells Fargo Alerts
Context
Alerts at Wells Fargo span dozens of alert types across multiple account categories, each backed by different systems and delivery rules. The existing experience was fragmented across destinations and difficult to navigate on mobile.
What broke at scale
-
Navigation reflected internal system boundaries rather than user intent
-
Users lacked a clear, complete view of what alerts were active
-
Edge cases multiplied as alert types and accounts increased
Design focus
I consolidated inbox and subscription management into a single destination, surfaced all relevant accounts by default and established reusable interaction patterns that could scale across alert types without custom handling.
Outcome
Users gained a clearer mental model of the system, while the design remained flexible enough to support dozens of alert types and future expansion without redesign.
E*TRADE Customer Service Intranet
Context
E*TRADE’s customer service teams relied on a patchwork of legacy systems, vendor tools and internal applications to support live customer interactions. Critical information was scattered across dozens of pages.
What broke at scale
-
Agents had to hunt across systems while customers waited
-
Redundant data created confusion about which information was authoritative
-
Adding new tools increased fragmentation rather than clarity
Design focus
I designed a start page that presented all the relevant customer information in one view. This organized complex, system-driven data into a structure that supported rapid comprehension and confident decisions.
Outcome
The new interface reduced cognitive load for agents, improved efficiency and provided a scalable foundation as new tools were added.