Notification System
Designed a notification system form from scratch. Reduced critical alert response time by 70%.
About the Project
InvestIT is an AI-powered automated trading platform.
Due to NDA restrictions, design details cannot be shared.
My Role
Sole designer. Built the notification system from scratch: research, problem definition, solution design, dev handoff.
The Problem
Users were constantly switching between multiple tabs to track activity. One dealer put it plainly: "I keep missing important trade alerts because they're buried in different places."
Approach
Shadowing sessions and interviews with dealers surfaced a consistent pattern: multiple tabs open simultaneously, manual order status checks every few minutes, personal reminders set for things the system should be surfacing.
Understand how notification works
Three critical categories drove the design:
Redefining the problem
How do we help users respond to critical alerts on time without adding noise to an already busy workflow?
The Solution
Early concept testing
This first design was intentionally basic. One view for all notifications. The goal was to watch how users interacted before adding structure.
Iteration 1
Users needed grouping and categorization. The volume was too high to process linearly.
Iteration 2
Grouping wasn't enough. The time it took to understand each notification was the next bottleneck.
Outcome
Critical alert response time dropped 70%. Users stopped switching tabs to track status.