Designing
a System
for Managing Attention
How I designed a notification system for a data-heavy product without interrupting user workflows
The Problem
In complex systems, notifications quickly become noise.
They:
- interrupt workflows
- lack structure
- compete for attention
- lose relevance over time
At the same time, users cannot afford to miss critical updates.
The problem was not how to design notifications.
It was how to design a system that decides what deserves attention.
Context
The product is a data-heavy environment where users:
- monitor processes
- analyze results
- manage workflows
- operate across multiple contexts
Work is continuous and cognitively demanding.
👉 Any interruption has a cost.
User Scenario
A user is working across multiple processes.
They need to:
- stay aware of system changes
- quickly identify what matters
- take action without losing context
This requires:
- persistent awareness
- structured information
- minimal disruption
Key Insight
Notifications are not messages.
They are a system layer that manages attention.
System Overview
The solution was designed as a multi-layered system, not a single component.
It consists of:
- Top navigation → global awareness
- Notification center (Control Hub) → structured workspace
- Views (All / Unread / Favourite) → attention filtering
Each layer serves a distinct role.
1. Entry Point — Top Navigation
Notifications are accessed through the top navigation.
Role:
- real-time signal
- global awareness
- lightweight interaction
Design decisions:
- always visible
- context-independent
- non-intrusive
A signal, not an interruption.
2. Notification Center — Control Hub
Notifications are expanded into a dedicated workspace.
Instead of overlays, they become structured data.
Structure:
- table-based layout
- clear columns (category, title, description, time, tags, actions)
- scalable for large volumes
Why it matters:
- supports scanning
- enables action
- reduces chaos
From alerts to structured information.
3. All Notifications — Full Context View
The default view provides complete system visibility.
Purpose:
- transparency
- access to full history
- understanding system activity
Features:
- all categories visible
- chronological structure
- tagging support
- direct actions
Everything the system knows.
4. Unread Notifications — Attention Layer
This view isolates what requires attention now.
Purpose:
- reduce noise
- highlight urgency
- support prioritization
Design details:
- visual emphasis on unread items
- hover and selection states
- consistent interaction model
What needs your attention now.
5. Favourite Notifications — User-Controlled Layer
Users can define their own layer of importance.
Purpose:
- save relevant items
- support long-term tracking
- reduce reliance on memory
Features:
- star-based system
- persistent visibility
- cross-context access
What you decided matters.
6. Tagging System
Tags provide additional structure and context.
They:
- enrich information
- support filtering
- improve scanability
👉 Notifications become easier to interpret and manage.
7. Awareness & State System
Each notification has a clear state:
- unread
- read
- favourited
- tagged
This allows users to:
- track importance
- manage attention
- return to relevant items
Design Decisions
Notifications as a system
Not isolated UI elements
Separation of concerns
Each view has a clear purpose
Action over information
Every notification supports a decision
Progressive disclosure
From signal → to context → to action
Non-intrusive awareness
Supporting flow instead of breaking it
Outcome
- reduced cognitive overload
- clearer prioritization
- faster reactions to system events
- scalable architecture for future growth
Design Principle
Awareness should be continuous —
not interruptive.
System Context
This notification system is part of a larger UX architecture including:
- top navigation (control layer)
- side navigation (system structure)
Together, they create a cohesive system that supports both orientation and decision-making.
Final Note
This project changed how I approach notifications.
From:
- UI elements
To:
- systems that manage attention
© Zofia Szuca 2024
Brand and product designer