이 포트폴리오는 PC 화면에 맞춰 만들었습니다.
PC에서 열어 주세요.

This portfolio is made for larger screens.
Please open it on a computer.

Back

[ Mar 2025 - Jul 2026 ]

Lync (mobile)

ROLE
Product Designer
Sole designer, end to end
TEAM
Product Owner
Engineers
PLATFORM
Mobile Application

Overview

*voyages, cargo, port costs, fuel, and documents

eLSA is an operational platform for tanker operators, managing charter parties. As the existing mobile app offered limited value and was rarely used, I redesigned key workflows for mobile and built a design system to help operators work away from their desks while keeping the product consistent as it scaled.

Background

Mobile couldn't be a smaller version of web.

Web uses its wide screen to show and compare many data points
at once. Mobile has to decide what matters first.

Business goal

01. Enable operators to continue working beyond their desks by bringing key web modules to mobile.

02. Build a design system that enables the product to scale consistently.

Discovery

“The away-from-desk moment.”

Mobile was primarily used outside the office to quickly check necessary information and complete tasks on the go,
rather than to explore and manage the full scope of work at a desk.

Core user flow. Trigger: outside the office, something needs attention. Check: open the module, find the record, check the details. If action is needed, edit what's urgent, then save and verify. If not, confirm and move on. Then done.

Findings

The audit showed inconsistent ways
of handling information.

I reviewed the existing app against the core user flow above.

See the research behind these findings

Research files · 2 documents

Click to open

Divergence 01

Similar records, different structures.
Users relearn the app on every screen.

Key Finding

Similar workflows needed shared structural and interaction patterns that could be applied consistently across the product.

Subtraction at scale 02

Every module needs a different amount of information.
One fixed layout doesn't fit.

Key Finding

A shared structure was needed—
one flexible enough to adapt the amount and combination of information to each module.

Ownerless state 03

States such as View, Edit, and saved/unsaved were
defined and represented differently across screens.

Key Finding

Product states needed to be defined first,
then connected to components so they
could be represented consistently.

Design System

I stopped designing screens
I started designing the system.

None of the three could be fixed one screen at a time.
I built a shared design system first, then rebuilt the app on top of it.

Click to open

Solution

Three fixes, one for each problem.

I rebuilt the screens behind each problem on top of the design system.

01

Identification

SOLVES DIVERGENCE

Tell records apart at a glance.

02

State & Action

SOLVES OWNERLESS STATE

See the state and the next step.

03

Clarity

SOLVES SUBTRACTION AT SCALE

Read information without decoding it.

Identification 01

Each row leads with what identifies the record.
The rest moves to the detail page.

Prioritised the strongest identifying information

The new list with the identifying details first
The old list

State & Action 02

The current state and the next action are shown on screen,
not hidden in gestures or colour.

01

Made the Edit action
explicit from View

02

Made colour-only states
explicit with filter chips

03

Added a View page
to confirm saved changes

Clarity 03

Icons and empty fields were replaced with named labels,
so information reads directly.

What changed

Repeated design decisions became shared rules.

WEBMOBILE
01New ModulesOne-off screensReusable structures
02StateInconsistentStandardised
03Web vs MobileSeparateShared + Adaptive
04HandoffScreen specsComponent specs
05WorkflowManual decisionsReusable rules

Design work moved from solving the same problems again to applying and updating shared rules across 45+ modules.

Final
Notes

From Screens to a System

As modules grew, small inconsistencies appeared across screens. Shared principles keep the product consistent, while each module adapts to its own information and workflow.

Mobile as System Validation

Applying the system to mobile exposed where platform-specific rules were needed. I refined component sizing, density, and state representation, using mobile constraints to validate and evolve the system itself.

A Repeatable Design Framework

By turning recurring design decisions into reusable rules, components, and documentation,

I created a scalable framework that could be consistently applied across new screens and modules.

What I Learned

Design System V1 became more than a component library. It became a foundation for making, communicating, and evolving design decisions consistently across the product.

Research files

Research. To define the tasks users needed to perform on mobile, I first looked at the situations in which they use it. Existing customer feedback had not distinguished web from mobile, so I looked at when users use mobile and why, then cross-referenced feedback from the Product Owner with customer input gathered during the web renewal. Audit. Based on the workflow defined through research, I reviewed the existing mobile experience end to end using usability heuristics. Three recurring issues emerged: Hierarchy, Visibility and Clarity. A spreadsheet lists each screen with its heuristic, severity, issue and category.

Design System

Step 1, Audit. There was no shared design system or component rule, so I audited the existing screens to identify repeated UI patterns, inconsistencies and opportunities for systemization. Includes a pattern audit board and a component inventory. Step 2, Foundation and Tokens. Recurring UI values became a shared token system: global tokens for typography and spacing, and colour tokens organised by semantic role. Light and Dark modes share the same structure. Step 3, Components and Naming. Recurring patterns across 45+ modules became shared components named by style and state, shown with a badge variant matrix and a new module composed from existing parts. Step 4, Architecture. I defined how components compose into reusable information units. For mobile, record identification came before information volume, with the rest moved into detail views.