The Trading & Operations teams at Snowy Hydro were relying on ageing Excel-based tools that were slow, fragile, and isolated. Forecasting required manual reconciliation across teams and made collaboration difficult. Through focus group sessions, we identified the need to replace these legacy workflows with a collaborative, real-time digital environment, leading to the creation of the Scheme Model Dashboard Interface, designed to centralise forecasting and streamline day-to-day operational planning.
As UI Lead and UX Designer, I established the digital design framework from the ground up, setting up all visual assets and screen templates, creating a scalable design system for future development, and building interactive prototypes in Figma to test layout efficiency, theme variations, and visual hierarchy. The process required close collaboration with multiple teams to translate complex, spreadsheet-based workflows into a unified, intuitive digital interface.

The teams had a clear vision of what the new Scheme Model Dashboard needed to achieve. Because multiple departments would rely on the tool, it had to support a range of data views, permissions, and workflows, all within a unified interface. My approach focused on translating these complex requirements into a flexible, scalable design that balanced precision with usability.
Light and dark modes were developed to support shift-based work patterns, reducing visual strain during overnight operations. Each mode was rigorously prototyped and tested to ensure clarity, accessibility, and consistency across all data layers. This iterative process helped establish design principles that could guide future digital transformations across the organisation.
To support cross-team collaboration, I designed a sandbox environment that allowed team members to manipulate and test data inputs in real time. This shared space enabled users to adjust and experiment with variables, encouraging clearer visibility of potential outcomes and fostering more informed decisions around inflow and forecasting adjustments.
The layout system is built on a 12-column grid with an 8px baseline, ensuring consistent alignment and readable rhythm across dense data views. Spacing follows a rem-based scale mapped to the 8px step, which keeps components visually balanced and prevents anti-aliasing on high-density displays.
Containers, gutters and in-component padding follow the same step system to preserve hierarchy at every level. This creates a predictable structure that scales, whether displaying summary cards or complex tables.

The typographic system was scaled with responsiveness in mind, focusing primarily on dashboards designed for larger screens (1920 × 1080). Since the team was already using Inter across other digital products, it was retained to maintain visual consistency and avoid introducing unnecessary complexity.
While the design foundations were optimised for desktop environments, scalable typography tokens and modular layouts were established to support future adaptability across tablet and mobile, ensuring flexibility without extending the initial project scope.

The colour tokens were named using Tailwind CSS conventions, as requested by the development team. This approach aligned well with my workflow and experience in front-end design systems, ensuring seamless handoff and consistency between design and code.
Although not required for this phase, I also defined error, warning and success states to future-proof the palette and provide a complete foundation for subsequent iterations.

I implemented a concise icon set focused solely on essential functions to maintain clarity and visual simplicity. The system was designed to evolve with future phases, allowing new icons to be introduced as the interface expands. Interactive components follow consistent padding ratios (horizontal × vertical = 2:1) and align to the 8px baseline grid for visual balance.
As the Scheme Model Dashboard is a data-driven environment rather than a traditional application, there was no need for extensive UI components such as input fields or complex button variants. The interface employs a restrained set of button styles, limited to core states and hierarchies, to keep interactions straightforward and consistent.

The final interface consolidates complex operational data into a clear, legible system, delivered in light and dark modes and supporting faster, more confident decision-making in day-to-day operations.



