CASE STUDY 02 · Construction & Government · Under NDA

Designing 100% Traceable Construction Progress for Public Infrastructure

How I designed a monitoring platform that replaces estimated progress with verified evidence, for office teams and field crews.

Client
Provincial government agency
My role
UI/UX Designer · Freelance
Timeline
Sep 2026 – Nov 2026
Platform
Web app · Android field app

Under NDA — product name, client and logos are hidden. Screen data is sample data.

Designing 100% Traceable Construction Progress for Public Infrastructure — product overview
01 · The context

Hundreds of projects a year, many roles, one chain of trust.

A provincial government agency manages hundreds of construction packages across multiple regions each budget year. Reported progress affects contractor payments, so it must be accurate and traceable at audit.

Contractors & supervisors

Report and verify work on site, often with mixed digital literacy and weak signal.

Contracting officers & admins

Approve reports, validate baselines, assign packages and act on critical contracts.

Leadership & auditors

Need an overview and a clear drill-down, rather than data entry.

  1. 01Contractor submits
  2. 02Supervisor verifies
  3. 03Officer approves & locks
02 · The problems

Why progress data could not be trusted.

Stakeholder interviews in the client’s requirement documents pointed to four problems. Each one became a design target.

PROBLEM 01

Progress estimated, not measured

Contractors reported round percentages such as 10% or 20%, without measured volume.

Risk: invalid payment data and losses discovered only at audit.

PROBLEM 02

Reporting scattered across tools

Daily, weekly and monthly reports moved through messaging apps, spreadsheets and shared drives.

Risk: slow recaps, copy-paste errors and untraceable evidence.

PROBLEM 03

Deviations detected too late

There was no early signal when a project fell behind its plan.

Risk: escalation for critical contracts started too late to recover.

PROBLEM 04

Previous tools were not adopted

Two previous systems failed because they were heavy and complicated, not for lack of features.

Risk: a third tool that people work around instead of using.

03 · My role & approach

Design the system, then every role on top of it.

What I was responsible for

  • A design system shared by the web and mobile apps
  • Role-specific flows, menus and interface design
  • Fast prototypes and weekly reviews with the client and PM
  • Developer handoff and implementation review

How I approached it

  1. 1.Map each role’s jobs from the requirement documents into flows and menus
  2. 2.Build a design system first so web and mobile share one language
  3. 3.Prototype fast with an AI-assisted design tool and review weekly
  4. 4.Hand off and review implementation against the design
Measured, not estimatedTasks before chartsColour, icon & labelBuilt for the field
Shared design system — status colours, icons and labels
04 · The solutions

Every decision traces back to a problem.

SOLUTION 01 · Daily field report

Replace estimates with measured volume and proof.

Progress was a guessed percentage, and previous tools were too complex for small contractors.

Daily field report
  1. 01Use four short steps: site conditions, resources, work progress and preview.
  2. 02Mark required fields and show inline errors.
  3. 03Capture volume per work item, show remaining quantity, and take evidence photos with location using the in-app camera.
  4. 04Preview exactly what the supervisor will see.

WHY IT WORKS

The percentage is calculated from evidence, so nobody has to type a number.

SOLUTION 02 · Rules in the flow

Make wrong data hard to enter.

Reporting rules that depend on trust alone do not reliably prove where and when work was reported.

Rules in the flow
  1. 01Select a package before reporting.
  2. 02Restrict package selection to a 150 m radius and show distances for out-of-range packages.
  3. 03Allow one report per day and show the current day’s reporting status.

WHY IT WORKS

Rules that relied on trust are now part of the interaction, so data arrives in a state the office can rely on.

SOLUTION 03 · Verification chain

One traceable path from field to approval.

Evidence needs to remain traceable from submission through verification and approval.

Verification chain
  1. 01Require supervisor on-site check-in and signature before verification.
  2. 02Show the officer the submission and verification trail, constraints and deadline.
  3. 03Approve and lock in one action.

WHY IT WORKS

Only verified, locked volume counts toward progress and payment.

SOLUTION 04 · Role-based dashboard

Start each day with tasks, not charts.

Office teams need to know which work needs their attention, not just see a collection of charts.

Role-based dashboard
  1. 01Use summary cards that reflect each role’s work.
  2. 02Group the to-do list with deadlines and direct actions.
  3. 03Place plan, actual and deviation side by side in the package table.

WHY IT WORKS

The first screen answers “what needs me today?”, which is the question each role actually has.

SOLUTION 05 · Early warning

Surface deviation before it becomes a crisis.

Late signals left teams with too little time to act on critical contracts.

Early warning
Early warning — escalation pipeline
  1. 01Show a deviation banner on packages past tolerance, together with the next scheduled step.
  2. 02Compare plan against verified progress using an S-curve.
  3. 03Group the escalation pipeline by handling stage, with overdue items first.

WHY IT WORKS

Deviation is visible where people already work, and every critical contract shows its next step.

05 · Validation & iteration

Iterating with the people building and using it.

The project is in active development. Designs move from weekly client reviews to developer handoff, and implemented screens are reviewed against the design system.

06 · Outcome & reflection

Designing for adoption, not just for features.

Designing for adoption, not just for features. The work brings a shared interface language, role-based workflows and evidence-led reporting into one monitoring platform.

Start from the job

Mapping each role’s jobs first made a large requirement set manageable.

Design for trust

The most valuable interactions are the ones that make wrong data hard to enter.

Iterate in the open

Weekly reviews and fast prototypes kept decisions concrete and visible to the client.

NEXT CASE STUDY

Streamlining Student Admissions for 50% Faster Completion