A platform collecting and presenting key healthcare data for patients to make informed decisions; collaborating with private providers, NHS Digital, and experts to enable and measure better patient outcomes and improved service performance.

Other UX performance impacts are:

• Improvements in search functionality to serve 20,000+ monthly visitors, and ~419,000 visitors annual users.
• Improved accessibility (legibility, visual land-marking, access to fees information) in practitioners' profile presentation, serving the 97-98% patient user base.
• ~50% reduction in data submission time with progressive disclosure forms enhancing data quality, and addressing user feedback about profile gaps.

Initial handover

From the beginning of my tenure, this project followed a typical iterative process of Discovery → Research → Ideation → Design → Validation (and repeat), yet delivering under UK's CMA deadlines and regulatory mandates required a less 'agile' approach.

Throughout these cycles, my purview was:

• Heuristic Analysis and competitor / marketing analysis of other healthcare information platforms, identifying UX benchmarks and WCAG accessibility compliance gaps, with a focus on web-based search interfaces.
• Fast-paced Discovery, Ideation, wire-framing with stakeholders and internal clients.
• Assisted all three Engagement Teams in User Research, focusing on coordinating effective research strategies and interpreting findings.
• Provider-facing Portal and Website: I delivered regular updates to ensure CMA compliance and coordinated a staged implementation involving multiple stakeholders.
• Design System Overhaul from Sketch to Figma while bringing UI consistency across multiple products and versions.

To service the aforementioned these family of digital products, I developed a single MUI/React Atomic Design System from the ground up. This shared library features nested, responsive components, tokenisation (e.g., a GDS-inspired type scale, color palettes, and tools for managing UX microcopy (e.g. Frontitude), as exemplified below).

Discovery phase

The data-gathering side of this service consisted in complex roles each with different user permissions (Practitioners, Medical Secretaries, Hospital Administrators, and Anaesthetists).

• User Permission matrix showing relationships (e.g., MedSecs administering multiple Practitioner profiles)
• Practitioner Onboarding with regulatory verification requirements
• Multi-stage publishing workflows for data validation and release

This analysis highlighted user needs and friction points across both platforms. The Portal in particular demanded a lot from users of varying tech literacy, while patients needed a smoother search experience.

Design Stage

Prototyping a simpler search

PHIN's website search functionality warranted an UX audit of the current feature and a redesign to reduce complexity. To better illustrate this solution, I coded a a prototype on Cursor, which I analyse in more dept here.

Data standardisation and form improvements

For a more consistent column ordering and information hierarchy across the Portal, I proposed and iterated with medical advisors a structure inspired by the journalistic 5Ws: First when, then who, what, and how. This established documentation for best practices in table design, reducing cognitive load across tables.

For practitioner's data submission, I designed a stepped "wizard" form with progressive disclosure. This received praise both from stakeholders and end users who found it dramatically improved their task completion time.

Visual design

With Management's support, I migrated from Sketch to Figma, which enabled a unified Master Library serving the aforementioned patient-facing website and the provider-oriented portal.

I also introduced a new type scale inspired by GOV.UK design standards, enhancing visual hierarchy and laying groundwork for future WCAG-compliant accessibility improvements. The challenge was creating a design system that catered to two different frontend frameworks (React/MUI and Svelte) while gradually bringing their visual languages closer together.