Doctolib — Case study
The New Care Experience
After years of rapid product expansion, the interface Doctolib’s healthcare professionals (HCPs) work in had become fragmented.
Our baseline: French software, 2020
Features were added one by one without a shared architecture. The result: duplicated flows, visual inconsistency and significant cognitive friction.
This slowed onboarding, hindered decision-making and made it hard to introduce new technologies such as AI or adaptive UI.
At its core, the product lacked a scalable, unified mental model: one clinicians could navigate intuitively and product teams could build on consistently.
Our Vision
We are building the interface for the future of healthcare: flexible and intuitive, so every professional can deliver the best care.
Layout 1 / 2
The Desk of Cards
At the heart of the NCE is a shared mental model we called “The Desk of Cards.”
The metaphor treats the clinician’s digital workspace as a desk, with information and tasks laid out as UI cards. Cards represent patient details, billing status, documents or upcoming tasks, each placed intentionally, the way a professional organizes paperwork on a physical desk.
The structure is organized into four views:
Health
Clinical data and patient status
Financial
Billing, insurance and reimbursement
Administrative
Practice settings, roles and documents
History
Longitudinal patient activity
Each view surfaces only the cards most relevant to the user’s context.
Cards adapt to the moment, displayed in three layout modes
Base, Contextual and Focus, depending on how deep the interaction goes.
Research and Validation
The New Care Experience builds on more than seven years of clinical research in France, which gave us deep knowledge of HCP workflows, pain points and expectations.
To go further, we built a multi-layered research model to validate, challenge and evolve our thinking as we designed:
In-house medical teams
We hired healthcare professionals into our feature teams, bringing first-hand medical insight into daily product conversations.
The Doctolib Community
To reduce internal bias, we held continuous conversations with thousands of users, validating assumptions in real time as we built.
Monthly “Speed Testing” events
Structured, rotating sessions in our Paris and Berlin offices paired community members with designers to test live concepts and prototypes.
Field research
We complemented in-house and lab testing with visits to urban and rural practices across France and Germany, for diverse, unbiased insights.
Speed Testing in our offices
We turned the Doctolib offices into a research space, with monthly “speed dating” sessions for medical secretaries and doctors.
How We Built It
Rearchitecting the clinician experience was more than a design initiative. It was a cross-functional transformation of how we research, align and build product at scale.
We began with a full UX audit based on Jobs to Be Done (JTBD). The framework helped us:
Identify the most critical workflows for healthcare professionals
Align design, research and product teams on a shared view of user needs
Measure KPIs consistently, with every success metric tied to a defined job
This gave us the clarity to rebuild around user outcomes, not features.
Jobs to Be Done audit — click to enlarge
To turn vision into action,
we went beyond alignment and built a collaborative delivery model.
We broke the New Care Experience into modular building blocks: cards, views, layout logic, navigation and system states. Each was assigned to a feature team, with shared ownership and clear contribution paths.
Instead of running NCE as a separate track, we embedded the work in team roadmaps. This required broad cross-functional buy-in, which I led through product reviews, strategy syncs and delivery planning with engineering and product leadership.
Measuring success
To make sure the transformation delivers real value to clinicians and the business, we defined KPIs tied to our JTBD framework:
Share of users transitioned to the New Care Experience
Reduction in time-to-complete for top workflows (as mapped to JTBD)
Task completion rates across redesigned flows
User feedback through in-product prompts and community touchpoints
Bug and support ticket volume for NCE-specific modules
Adoption of shared components and navigation across squads
