AWS Monitron — Industrial Machine Health
When I joined the Monitron initiative, I discovered the core issue wasn’t the sensors — it was the mental model technicians relied on to diagnose machines. My work reframed the entire product around technician behavior, transforming Monitron from a sensor-first experience into a machine-first diagnostic tool that aligned with how maintenance teams actually work.
Advanced UX Designer
January 2023 - March 2024
PM, Engineering, Content
Web (Industrial IoT)
Why This Problem Mattered
For industrial environments, clarity isn’t a luxury — it’s a safeguard.
The existing experience caused:
frequent misassignments during setup
uncertainty about what machine needed attention
delayed reaction times to anomalies
growing distrust in the insights generated
And at scale (hundreds or thousands of machines), this wasn’t just a UX issue —
it was a risk to uptime, safety, and operational cost.
My responsibility as Principal Designer was to create a system that technicians could rely on instantly, even in noisy, high-pressure, high-stakes environments.
The Insight That Changed Everything
After interviewing technicians, walking job sites, and mapping their workflows, one pattern emerged repeatedly:
“Don’t make me think about the sensor.
Help me understand the machine.”
This led to the central reframing of the entire redesign:
Shift from a data-led interface to an operationally-led interface.
Instead of asking the user to interpret raw vibration/temperature data,
the system needed to communicate:
Is this machine healthy?
What changed?
What should I do next?
What evidence supports that recommendation?
This reframing served as the design foundation.
35%
Improved onboarding process
25%
Increase in user retention
84%
Increase in time spent on website
My Design Approach
Simplify the mental model: I restructured the hierarchy around:
Machine → Sub-assembly → Component → Position → Sensor
This aligned the UI with how technicians physically approach equipment.
Guide technicians through complexity: Key UX decisions focused on clarity under pressure:
A machine overview that consolidated all insights into a single at-a-glance view
Clear “current condition” status with supporting evidence
A chronological machine health timeline to show cause-and-effect patterns
Sensor assignment flows with guardrails to prevent misconfiguration
Predictive insights surfaced only when confidence thresholds were met
Every component of the interface needed to “speak the same language” of real-world maintenance workflows.
Design for industrial realities: Environmental constraints shaped UI choices:
Glove-safe tap targets
Reduced cognitive load in noisy environments
Hierarchical breadcrumbs for quick navigation
Color and icon semantics tuned for abnormal conditions
Clear differentiation between degraded, warning, and critical states
This ensured Monitron remained usable in the field — not just at a desk.
The New Experience
Here’s how the narrative flow changed for the user:
Before
“Which sensor is this? What machine is it attached to? Is this an issue?”
After
“The machine is in a warning state. Here’s what changed, why, and what to do next.”
Outcome & Measurable Impact
Even before full rollout, early testing validated our direction:
45% faster sensor assignment
Considerably fewer misconfigurations during onboarding
Technicians reported immediate clarity on machine vs sensor state
Improved trust in predictive insights
Increased adoption of machine analytics features
But the biggest outcome was qualitative:
“Now I don’t have to ask myself what the app is telling me.
It matches what I see on the floor.”
That alignment between digital understanding and physical reality is the true measure of a successful industrial UX system.
Business Value Delivered
The redesign contributed to:
Lower downtime risk
Faster anomaly response
Reduced need for technical training
Higher customer satisfaction
Stronger enterprise adoption for Monitron
Clearer upgrade path for multi-site deployments
This wasn’t just UI improvement —
it was strategic product acceleration.