Type of product
Internal government digital platforms serving multiple departments.
Stage
0→1 — Design System created from scratch to support current and future products.
Constraints
• Legal and regulatory requirements
• Fixed political deadlines
• No usage analytics or historical data
• Multiple teams working in parallel
• No existing design governance
• Interviews with internal teams to understand recurring patterns and pain points
• Documentary and regulatory research to identify non-negotiable constraints
• Accessibility and contrast analysis focused on green-based palettes requested by stakeholders
• Ongoing alignment with development teams to validate components incrementally
This project involved a clear tension between accessibility requirements and strong stakeholder preferences for a clean, light green-based color palette.
While I initially proposed multiple color palettes that fully met WCAG contrast standards, stakeholders prioritized visual cleanliness over strict accessibility compliance. The proposed palette introduced risks for users with low vision and color blindness, particularly in critical health-related flows.
Instead of forcing a binary decision, I reframed the discussion around risk and impact. I negotiated a compromise: adopting a visually clean palette aligned with stakeholder expectations, while enforcing higher contrast standards for key components, critical actions, and informational feedback.
This approach balanced political constraints with user needs, ensuring acceptable accessibility levels in high-impact areas without blocking adoption of the Design System.
While the Design System was still being built, parallel product teams continued designing independently, frequently introducing new components without alignment. This created a clear risk of fragmentation — exactly the problem the Design System was meant to solve.
Enforcing strict rules too early would slow delivery and reduce team buy-in. Ignoring the issue would compromise consistency at scale. I chose a middle path: lightweight governance through visibility and shared ownership.
I established short, recurring alignment sessions to share Design System progress, review active product work, and align on component decisions in real time. This created a feedback loop where teams felt heard while gradually adopting shared standards.
As a result, teams transitioned toward a unified visual language without blocking ongoing delivery, and the Design System became a living reference rather than a rigid constraint.
• Unified visual language across multiple products
• Reduced UI inconsistency and duplicated design decisions
• Faster alignment between design and development teams
• A shared foundation that enabled new products to be designed more efficiently
If I were to measure impact, I would track navigation efficiency, reduction in rework, and time-to-delivery across products using the Design System.
This project reinforced that early-stage products are shaped less by ideal solutions and more by decision-making under constraints. Technical knowledge alone is not enough — progress depends on navigating stakeholder preferences, timing conversations correctly, and managing risk without blocking momentum.
I learned how to establish structure while building something new, guiding teams through shared rules without slowing delivery. More importantly, I understood that successful design in early-stage environments is about creating alignment, not control — and making imperfect decisions that allow products to scale sustainably.