Lillian Ip-Koon - Macmillan Orders

Macmillan Orders

Macmillan Cancer Support

Macmillan Orders is a service platform that provides free cancer information and merchandise to the public, supporters, healthcare professionals, and staff. During a major migration to a new platform, I led the user-centred design strategy, built the design system, and crafted UX writing to create a more consistent, scalable, and accessible service.

The platform launched successfully in October 2024 and attracted over 3,000 users in its first month. The redesigned experience improved navigation, product discovery, and the ordering journey, while also streamlining order management for the supporter care team. By aligning the service more closely with Macmillan's brand and introducing a shared design system, the work improved both the user experience and the organisation's ability to deliver efficiently at scale.

Case Study 1:
Improving product discovery

I led discovery work to understand why users struggled to find products within the Macmillan Orders service. Through workshops with users and subject matter experts, we identified key issues: an ineffective search function, fragmented information architecture, and a tree structure that depended heavily on prior user knowledge. These problems created significant barriers for both healthcare professionals and public users trying to access essential materials.

Two site map diagrams, made of collections of post it notes

Wireframes of a two web pages, with post its around it

Designing a clearer information architecture

To address this, I facilitated co-design workshops with stakeholders across multiple departments to rethink the content structure from the ground up. This resulted in a complete reorganisation of the information architecture, supported by low-fidelity wireframes and structural diagrams that clarified navigation pathways. These artefacts informed high-fidelity prototypes, where I also contributed UX writing to fill gaps and improve clarity. As there was no dedicated content designer on the team, I proactively brought one in at a later stage to review and strengthen the content.

Proving impact through user insight

I validated the new approach through user testing with two key groups—professionals and public users—using UserZoom. The results showed a clear improvement in content discoverability and ease of navigation. After launch, I established ongoing feedback loops through user sessions and implemented Hotjar to monitor behavioural metrics such as rage clicks and drop-offs. This ensured continuous optimisation based on real user behaviour rather than assumptions.

Screenshot of the usertesting platform, userzoom, showing data from a piece of research

A user journey of a Macmillan supporter user type who is donating money, with painpoints highlighted

Case Study 2:
Identifying process inefficiencies

I identified a critical point of friction in the supporter ordering journey: a manual, phone-based approval process that created unnecessary effort for both users and staff. Taking a user-centred approach, I investigated the problem by working closely with supporting teams to understand operational realities. This revealed recurring issues—supporters often misunderstood ordering criteria, requested more items than needed, and struggled with the requirement to manually note product codes before placing orders.

To better understand the experience, I mapped the end-to-end supporter journey, which exposed a clear service gap. Users were unable to complete orders independently, resulting in avoidable delays and a heavy reliance on staff intervention. This not only impacted user satisfaction but also created a significant operational burden, making a strong case for redesigning the process.

Designing an automated approval flow

I collaborated with the product owner and developers to explore how existing platform functionality could support a more efficient, automated approval process. I redesigned the checkout journey to include targeted questions that captured the necessary approval information upfront, reducing the need for manual intervention. Throughout this process, I worked with an accessibility specialist to ensure the solution met WCAG 2.2 standards. Usability testing with a high-fidelity Figma prototype via UserZoom confirmed that the questions were clear and supported successful task completion. I also identified a wider service misalignment that routed some users incorrectly, and facilitated cross-team discussions to agree a pragmatic compromise that reduced user impact within operational constraints.

Screenshots of a prototype of the shopping cart page on Macmillan Orders website

Iterating based on live performance

Following the release in early 2025, I monitored live performance and defined success measures, including rejection rates and support queries. Initial data showed that 43% of orders were still being rejected due to missing information. I traced this to backend limitations and worked with frontend colleagues to redesign the checkout flow in a way that aligned with technical constraints while minimising development effort. The updated solution introduced mandatory fields and clearer messaging to set expectations and improve data quality. After implementation, rejected orders significantly decreased, demonstrating the value of continuous, data-driven iteration.

Case study 3:
Establishing a scalable design system

I initiated the rebuild of Macmillan’s design system, which was previously outdated and functioned more as a static style guide than a reusable system. I introduced Figma as the central tool and established a clear framework for how the system would be structured and maintained. As interim lead, I upskilled the team by arranging external training on design tokens and supporting designers in applying this knowledge in practice. I also mentored junior designers to build confidence in Figma, aligning the team around consistent ways of working while enabling individuals to contribute to areas of the system where they had the most expertise.

Screenshot of accessibility requirements and design tokens in tables in a figma file

Screenshot of a figma file with information about cards and components in the design system

Embedding and evolving the system

Through the Macmillan Orders project and two other services I was leading, I built and embedded the design system in parallel with live delivery, ensuring components were tested in real contexts and adapted to different technical constraints. This led to a more consistent experience across Macmillan’s digital services, faster delivery as teams reused shared components, and fewer bespoke UIs being created from scratch. The system also established clearer standards for layout, interaction, and accessibility, which helped reduce fragmented design patterns and improve recognisability for users. All work streams adopted the design system as soon as it was created, improving consistency as we continued to move forward with a small team of 3–4 designers for the entire organisation. By influencing the engineering lead to support the design system, I secured dedicated developer time to maintain components and host regular design–developer forums, helping treat the system as a shared asset embedded in both design and development workflows.

Key outcomes

Improved product discovery
Restructured the information architecture and navigation, making content easier to find for both public and professional users.
Reduced operational burden
Transformed a manual, phone-based approval process into a more efficient, partially automated journey, reducing staff intervention.
Increased order success rates
Iterated on the checkout flow using live data, significantly reducing rejected orders by improving data capture and clarity.
Established a scalable design system
Rebuilt the design system in Figma with reusable components, improving consistency and speeding up delivery across multiple projects.
Embedded continuous improvement
Introduced ongoing user feedback loops and performance monitoring (e.g. Hotjar), enabling data-driven iteration beyond launch.

Reflection

This project demonstrates my ability to work across the full end-to-end lifecycle—from discovery through to live service optimisation—while balancing user needs, operational realities, and technical constraints. A key strength was identifying and addressing not just interface issues, but underlying service gaps, particularly in the ordering and approval flow.

Leading across multiple workstreams also highlighted the importance of adaptability. In the absence of dedicated roles such as content design, I stepped in to fill gaps while still bringing in specialist input where possible. Similarly, building the design system alongside live delivery ensured immediate value, but with hindsight, introducing clearer governance and measuring adoption earlier would have strengthened its long-term impact.

Overall, this work reflects a shift from delivering individual improvements to shaping more coherent, scalable service experiences—while continuously using evidence and collaboration to refine decisions.