I led the 0-1 design of a Project Planner application, addressing the fragmented project and labor tracking system across Engineering Departments at EMCOR.
It improved project management with standardized logging, streamlined communication, project schedules, & more!
Problem Statement
"How might we standardize project and labor tracking for EMCOR's Engineering departments to improve efficiency?"
Disparate project tracking methods led to mismanaged data, as each department struggled with logging 100+ projects.
Research & Users
I centered research on the Irvine branch.
Artifact Audit: 55% (87/158) of projects had missing details, like logged hours.
Contextual Interview: a group interview (1 manager, 2 engineers) regarding their current system showed a satisfaction score of ~52%.
I prioritized insight from this leading branch to build a scalable base solution for future department rollout.
"There's too much human input & error"
- Engineering Manager
"It's not automated & takes too much time to communicate what's left to do!" - Engineers
Synthesis
Affinity mapping confirmed key themes from my research.
Data fragmentation, clunky workflows, manual errors, and poor security were large pains.
A critical insight emerged: Managers and Engineers had distinct differences, so I created "mini profiles" to define these role-based behaviors, needs, and goals.
GOALS
Eliminate Data Fragmentation
Improve Workflow Efficiency
Increase User Satisfaction
Ensure Accurate Tracking
Provide Role-Based Controls
⬇️
FEATURES
Search & Filter
Secure Role-Based Access
Log & Track Labor Hours
Task Interactions/Reminders
Communication Tools
Project Creation & Editing
Subtask Creation & Management
Gantt Chart (Data Specialist delegation)
Ideation & Clarification - IA
In my initial sitemap, I struggled with visualizing how logged hours impacted project budgets differently for Managers and Engineers.
Representing complex financial relationships while separating user goals was hard!
NOTE: My confusion!
I conducted a targeted follow-up interview, moving beyond observation to co-analyze existing data.
My team helped me visualize the flow on whiteboard <3
By dissecting real-world edge cases (like over budgets) together, I finalized an IA that clarified these complex financial-logging connections and distinct user paths.
Sketches & Design System
I used low-fidelity sketches purely for content blocking because I previously created and managed EMCOR's comprehensive design system.
This approach was my efficient way of streamlining the eventual development to high-fidelity as components and behaviors had already been defined.
Low-Fi Prototype & Testing Transition
I created a low-fidelity prototype in Figma. It's hi-fi look was my strategy to introduce a crucial fidelity testing phase to the department.
Previously, my department had a costly practice of developing applications directly from sketches. By focusing user feedback on core functionality and workflows pre-build, I accelerated validation and lessened wasted resources.
A high visual standard met users halfway in expectations and minimized their known struggles with abstraction, enabling them to focus solely on feature validation and usability.
Lo-Fi Testing & Findings
I ran a structured group usability test.
Due to time constraints and network limitations requiring real-time simulation instead of Figma access.
Tasks were segmented by user role (Manager, Engineer, vs. All) to gather specific feedback in a shared setting.
This allowed me to refine core information displays and form fields, while also surfacing new issues and opportunities for for process improvements I previously overlooked.
Hi-Fi Prototype
Leveraging a rapid prototyping method, I built the application's front-end in Power Apps.
This allowed for immediate integration of low-fidelity testing feedback into a fully playable, high-fidelity experience, enabling faster validation and iteration cycles.
UI Iteration Example
During lo-fi testing on the Hours Log page, Engineers attempting to search and filter their timesheet entries immediately struggled with the existing date filtering.
They wanted highly specific control, "This selection seems arbitrary — why a broad time range? I know the exact dates I want".
I integrated this insight to the hi-fi prototype, replacing the static date dropdown filter with a calendar picker for "to-from" dates.
This would match their specific mental model and enable efficient, precise data entry.
Hi-Fi Testing & Findings
I led a structured, final validation testing session with the Engineering team.
1:1 playtests were administered. We observed users as they navigated the live-test application on their laptops.
This phase shifted focus from core usability to identifying edge-case refinements, visual polish, and back-end data consistency issues.
The test confirmed the UI iteration was a success, eliminating the confusion and stop-points users previously experienced during their hour-logging exploration.
Launch & Continuous Improvement
Following iteration, the Project Planner deployed with Irvine Engineering's live data.
I established a continuous improvement workflow, utilizing PowerApps analytics and live monitoring to track key performance indicators (KPIs).
Ex) task-completion times and user engagement.
Quantitatively, this data was paired with a qualitative feedback loop — an in-app error report form and follow-up interviews — to address performance issues and user pain points.
This built a strong user relationship and enabled the product team to implement continuous feature improvements in 1-2 week rapid prototyping cycles.
Impact & Future
Post-launch surveys confirmed a rise in user satisfaction from 52% to 85%.
46% (72/158) of previous projects had missing labor hours — now, 100% of new & active projects have labor information
~5x faster data logging
Irvine's project tracking systems condensed from 8 methods to 1
The Project Planner fundamentally transformed project management by standardizing fragmented tracking methods into one centralized, role-based solution, proving its value and establishing a clear path to scale the product to other branches.