itero
Back to Work
Industrial IoTService DesignPredictive Analytics

Smart Condition Monitoring & Predictive Maintenance

Transforming a research-phase sensor trial into a full-scale predictive maintenance platform — reducing maintenance costs by 61% for industrial operators across 5 continents.

ClientGlobal Industrial Lubricants Company
ScopeService Design → UX/UI → Product Strategy → Analytics
SectorsPower · Off-Highway · Marine · Mining
Predictive maintenance monitoring dashboard
61%

Reduced maintenance costs

45%

Machine defects addressed through better monitoring

62%

Improved oil management, reduced unplanned downtime

Situation

Sensor data trapped in a lab prototype

A global industrial company had developed IoT sensor technology that could monitor lubricant oil condition in real-time across heavy machinery — engines, generators, haul trucks, marine vessels. The sensor hardware worked. The research data was promising. But the technology was trapped in a lab-phase prototype with a handful of pilot customers and no scalable digital experience around it.

The existing solution was a bare-bones MVP driven entirely by the R&D team's assumptions about what users needed. No user research. No consideration for different roles. No path to becoming a commercial product.

They needed to turn a sensor trial into a service that industrial companies would actually pay for and use daily.

Challenge

Three problems, one platform

Multiple user types, conflicting needs

Four distinct roles — maintenance engineers, technical advisors, engineering managers, and operations managers — each with fundamentally different priorities, workflows, and information density requirements.

Industry-agnostic, context-specific

The platform needed to work across Power, Mining, Marine, and more — each with different equipment, sensors, and norms. A single configurable architecture that adapts without custom development per industry.

From data display to decision support

Translating raw sensor readings into actionable decisions: "change this oil now," "schedule maintenance next week," "this equipment is fine." A fundamental reframe from technology push to customer pull.

Our Approach

Applying the Itero Loop

We applied our four-phase iterative methodology to evolve the service from sensor trial to scaled product.

Phase 1

Understand

We interviewed maintenance engineers, technical advisors, engineering managers, and operations managers across multiple sites and industries. The goal wasn't to validate the existing MVP — it was to understand how each role actually made decisions about equipment maintenance.

Key insight: operators were using time-based maintenance schedules (change oil every X hours regardless of condition). The platform's real value was enabling the shift from time-based to health-based maintenance.

Phase 2

Shape

We designed four distinct user experiences tailored to how each role thinks and works:

  • Maintenance Engineers — Glanceable equipment health with clear action recommendations.
  • Technical Advisors — Deep trend analysis with oil property graphs and data export.
  • Engineering Managers — Site-level cost tracking and productivity dashboards.
  • Operations Managers — Multi-site bird's-eye view across 20+ locations.

For the industry-agnostic challenge, we designed a smart backend mapping layer — assets, equipment types, sensors, and configurations abstracted so the same UI adapts across verticals without custom development.

Phase 3

Build

We integrated usability testing into every agile sprint. Key deliverables included a live monitoring system translating sensor data into four clear states with actionable recommendations, a remaining oil life prediction engine with supporting trend visualizations, a real-time equipment health dashboard with filter, group, sort, and pin functionality, a self-service asset onboarding system reducing vendor dependency, and customizable reporting across live snapshots, historical analysis, and raw data exports.

Phase 4

Refine

We embedded analytics from the start — tracking which pages each user type visited, which features drove engagement, and how behavior differed by role, organization, and industry. This created a data-driven decision culture where every design choice was backed by evidence. Continuous remote usability testing across sprints generated a structured findings log that shaped every iteration.

Results

From pilot to global product

The platform transformed from a research-phase sensor trial into a fully commercialized, multi-industry digital product deployed across 5 continents.

61%

Customers reporting reduced maintenance costs

45%

Machine defects addressed through condition monitoring

62%

Better oil management, reduced unplanned downtime

5+

Industrial verticals served from a single platform

4

Distinct role-based user experiences

Time → Health

Shifted entire customer base to health-based maintenance

Service DesignIndustrial IoTMulti-User Type UXDashboard DesignPredictive AnalyticsResponsive Web AppUser ResearchUsability TestingData VisualizationDesign-Led EngineeringAnalytics Implementation

Have a similar challenge?

We'd love to hear about it.

Start a conversation