Redesigned the legacy messaging system to improve user experience and increase engagement while introducing modern messaging functionalities.
Led the full design process from ideation to implementation, strategy and execution while aligning cross-functional teams towards the same goal.
Conducted user research, user flows and competitive analysis to map pain points in the existing messaging flow.
Restructured the message hierarchy to reduce cognitive load and quickly set the most relevant content first.
Built interactive Figma low-fi prototypes and ran two rounds of usability testing with non-users and team members iterating on findings between each session.
Designed a modern, accessible messaging interface aligned with SACEM's design system, introducing new functionalities such as rich-text replies, attachments and improved message organization tools.
SACEM is the French society responsible for collecting and managing the royalties for music creators and publishers. We have been their product partner since 2020,
providing them with a dedicated platform with perks ranging from music industry discounts and event agendas to member profiles, networking, second-hand equipment marketplace...
Ahead of a major event that required using the platform messaging system, the client sat with us to share their requests which led to us redesigning the whole messaging system to
a much more modern one. As the lead and only designer for this project, I took part in the very first calls with our client manager and SACEM's team to understand their needs before any brief existed. All client
communication ran though our account manager, which meant translating business needs carefully and sharing them with the IT team quite early to define the whole project.
The project came with a tight deadline: a full messaging redesign had to ship before a major client event, roughly two months from kick-off. This forced early and difficult scope decisions, with several features cut before a single wireframe was drawn.
The platform itself added a difficult task. SACEM ran on a CakePHP version that was nearly a decade old, which put several modern messaging such as real-time updates, inline media previews, threaded replies, impossible to accomplish. We had to find alternative approaches that delivered equivalent user value that our technology could actually support.
Scheduling was also a problem we encountered. The project ran through August to October, overlapping with summer holidays on both the team and client side. Feedback sessions were hard to coordinate, and at times we had to move forward on key decisions without direct client validation therefore relying on our research in order to advance.
Usage data from the existing system set the current status of the messaging very clear: the average conversation contained just 2.8 messages. Users weren't continuing threads for two main reasons, they were quite limited with what they could do or share within the legacy system, and perhaps because they had quietly switched to more convenient channels like email, whatsapp or other messaging tools.
With those key pain points mapped, the objectives were clear. The redesign needed to make conversations more likely to continue by enhancing the legacy system to a more up-to-date one.
That meant introducing message organisation tools: such as being able to mark messages as read or unread, archive threads, and/or flag important ones as favourites, so users could manage their inbox as one would expect. We also prioritised having a modern reply experience: formatted text and attachment support and reworked the visual hierarchy so unread and high-priority messages were immediately scannable in a quick look.
Every design decision was filtered through two lenses: does it work within the existing technology, and can it ship within the expected timeline.
Screenshots of the old messaging system
Before defining the feature, I spent time auditing existing messaging systems, in both user apps and enterprise tools to stress-test our requirements list. The goal was to confirm that every feature we were proposing was something realistic, and that we weren't accidentally inventing something that would feel unfamiliar or require users to re-learn basic behaviours.
The new system needed to feel immediately recognisable, close to what people already used in their daily lives so that the learning curve was would be zero for most.
Conducted interviews and surveys to understand user needs and pain points.
Created low-fidelity designs and got buy-in from stakeholders.
Developed high-fidelity prototypes for user testing and validation.
Tested with real users to validate design decisions and gather feedback.
Refined the design based on testing results and stakeholder input.
Collaborated with developers to ensure accurate implementation.
The final design looked quite different from where we started. Early concepts had to be revisited and some ideas were cut due to technical issues or impossibility, others evolved through feedback from the client and testing. Both the client and our side agreed it hit the right balance: a modern, usable messaging experience that fit within the platform's existing architecture.
Feature 1 — New message threading view
Feature 2 — Notification centre redesign
+130% monthly messages — launched ahead of a major platform event, monthly messages grew from ~3k to ~7k in the first quarter post-launch.
Event success drove platform growth — the music collaboration event, co-led by the client and a well-known producer (Joachim Garraud), was a success. The redesigned messaging played a role in that experience, earning trust from the organiser and participants, bringing new users to the platform and generating positive word-of-mouth.
Sharing early concepts with the IT team before actually investing time in even low-fi design consistently saved a lot of work. The sooner that possibility is discussed, the fewer decisions that have to be undone.
We encountered real issues mid-project since legacy technology couldn't support certain interactions we had designed and early discussed. It was a good reminder that good design needs to keep in mind what's achievable, both in time and in code.