Case Study: Avigilon Cloud
AI-Powered System Health Dashboard

Overview

As Director of Design & Research at Motorola Solutions, I led the end-to-end design strategy for Advanced System Health, a key capability within the Avigilon platform which supports cameras for all products at Motorola Solutions. The Cloud Advanced System Health was designed to deliver centralized, real-time visibility and control for integrators, IT administrators, and security operators managing thousands of devices across global locations.

My Role

  • Title: Director, UX Design – Cloud Video & System Intelligence

  • Team: 5 designers, 1 researcher, 1 PM, 2 engineering leads + their teams

  • Timeframe: 9 months (concept to rollout)

  • Key Partners: Product, Engineering, Support Ops, Sales Engineering, and major enterprise clients (e.g., education, healthcare)

  • Tools: MUI Charts, D3.js, Microsoft Clarity

The Problem:
Eroding trust in the product

Many of our enterprise customers rely on hundreds—or even thousands—of IP-connected security cameras across campuses, warehouses, or stores. As security systems scale to thousands of sites, servers, and cameras, finding and managing health and uptime becomes unmanageable. Customers struggled with disconnected and misleading device statuses, lack of actionable insights, and repetitive troubleshooting, leading to eroding trust in the product.

The Solution:
Centralized AI Health Monitoring

A centralized health dashboard and AI tagging to locate cameras and manage their health automatically by providing a "single pane of glass" that summarizes health across all cloud-connected Avigilon sites, servers, and cameras, with actionable insights and immediate intervention capabilities for enterprise-scale deployments. The design is custom and responsive, built to meet the needs of all viewports — mobile, tablet, and desktop.

Approach

Discovery

Identify key pain points and opportunities around how admins monitor, interpret, and act on system health issues across Avigilon products.

  • Internal collaboration conducted by extensive interviews and card-sorting exercises with product management and sales engineers, and other internal subject matter experts to understand what "system health" meant at scale.

  • Contextual interviews with IT managers, system admins, and security directors in verticals like higher education, logistics, and enterprise retail

  • Job shadowing with support engineers handling frequent tickets

  • Support case analysis to identify the most common and costly system failures

  • Journey mapping of diagnostic workflows and escalation paths

Competitive Analysis

Benchmarked top competitors and on-premises solutions (Genetec, Milestone) to surface best practices and pain points around status observability and dashboard visualizations. This analysis revealed common issues with information overload and lack of actionable insights.

Design Thinking Workshops & Early Insights:

Affinity Map Clusters

Visibility & Awareness

“I don’t know something’s wrong until a customer calls.”
“Health info is buried in different dashboards.”
“We need one place to see system uptime and camera status.”

Theme: Lack of unified visibility and fragmented data.
Opportunity: Build a single-pane system health view with real-time status.

Prioritization & Signal Clarity

“Too many alerts — I can’t tell what’s critical.”
“Low battery looks the same as a full outage.”
“We need color coding or severity levels.”

Theme: High noise, low signal clarity.
Opportunity: Introduce smart alerting with severity tiers, context, and relevance filters.

Actionability & Resolution

“It says camera offline, but what do I do next?”
“I waste time guessing what caused the issue.”
“We need troubleshooting steps right there.”

Theme: Alerts don’t translate into action.
Opportunity: Create guided resolution flows and AI-assisted root cause suggestions.

Trust & Transparency

“Sometimes the system says a camera’s offline when it’s not.”
“False alerts make me stop trusting the dashboard.”
“I need to see why it flagged this issue.”

Theme: Erosion of trust in system reliability.
Opportunity: Provide explainable diagnostics — show data source, timestamp, and confidence level for each alert.

Predictive & Proactive Health

“We always fix after it breaks.”
“Would love alerts before a failure — like overheating warnings.”
“AI should tell us what might fail next.”

Theme: Reactive vs proactive system management.
Opportunity: Develop predictive maintenance models and early warning insights using telemetry data.

Key Insight

  • Admins didn’t want more data — they wanted clarity, confidence, and control.

  • System health needed to evolve from status reporting to situational intelligence.


    This initial discovery gave us strong initial hypotheses

Prototyping

Led rapid prototyping using real data to validate design concepts — exploring D3 sunburst and other data-dense visualizations. While initial layouts were modeled in Figma, we advanced to interactive prototypes built with D3 and JSON to test component behavior, interactivity, and performance with live datasets.

Optimizing Camera Operations with AI: Health Monitoring and Scene-Based Tagging

Camera Health, an AI-powered feature to identify camera image quality and recording deficiencies. 
Camera Tags helps user easily find cameras based on scene descriptions driven by AI.

Testing &Validation

Performance Testing
Partnered with engineering to stress-test the dashboard using data factory scripts, simulating enterprise-scale deployments with thousands of devices and real-time updates.

User Feedback Integration
Integrated Microsoft Clarity for analytics and heatmapping in CI to inform continuous design iteration before production launch.

Customer Validation
Conducted validation sessions with major enterprise clients, confirming alignment with key drivers for cloud adoption over on-prem solutions.

Outcomes & Impact

  • 50% reduction in support tickets for health-related incidents

  • 40% faster resolution times reported by key enterprise clients

  • Used research insights to inform roadmap, not just polish the UI

  • Featured in EPF demos as a critical advancement in operational confidence

Leadership & Learnings

  • Brought design, product, engineering and support together in co-creation workshops

  • Elevated research from “early-stage input” to a continuous strategy tool. We introduced Microsoft Clarity.

  • Drove accessibility and clarity across all views (high-contrast design, assistive labels,  built-in help, etc.)

  • Created a framework for closing the research loop—a practice now replicated across Avigilon Cloud projects

What’s Next

  • Extending system health to include:

    • AI analytics readiness

    • Storage utilization forecasting

    • Mobile alerts for on-site integrators

  • Creating integrator dashboards for multi-site fleet health

  • Building auto-resolve logic using machine learning from recurring issue patterns