Probando GmbH
Two product redesigns revealed the same system problem.
I joined Probando to improve its internal dashboard. The public website came next. In both projects, the same issue slowed the work: the existing design system was hard to adapt and unclear for engineering. I rebuilt it with variables, tokens, reusable components and documentation, then used it to design Probando’s payment system from zero.
Impact
- 3Products using the same design systemThe internal dashboard, public website and ProbandoPay shared components, variables, tokens and documentation.
- 100+Figma components rebuilt and addedOptimised and built states, usage guidance and Storybook references made the library easier to use across design and engineering.
- 0 → 1Payment product designI defined connected journeys for participants, study sites and clients, from onboarding to payout reporting.
- 0Workarounds reported after the dashboard redesignThe internal team reported that the workarounds which had become part of daily use were gone.
Overview
Context
Probando is a European health-tech startup for clinical trial recruitment, management and participant payments.
Initial tasks
Improve the internal dashboard, then redesign the public website.
Problem I found
The existing design system was rigid, difficult to adapt and not clear enough to guide engineering.
My response
I rebuilt and extended it with variables, tokens, reusable components, documented states and Storybook references, then designed ProbandoPay on it from zero.
Result
The dashboard, website and payment product run on one design system now.
The design system was never on the brief; it became necessary while I redesigned the first two products.
Project sequence
01 · Internal dashboard
Only a few employees could use the dashboard without workarounds.
The brief was to improve the internal dashboard. The deeper problem was operational dependence.
- ~100Active trials at the time
- A fewEmployees could complete key tasks independently
- Workarounds removedReported by the internal team
Knowledge was stored in people instead of the product.
The dependence
A few power users knew how to complete tasks, new employees needed their help, and questions often became requests to engineering.
The real constraint
For a growing team, this made onboarding and daily delivery depend on individual availability.
The constraint was not only usability.
It was the number of studies the team could operate without adding more manual support.
Decision
I reorganised the dashboard around the study.
The product structure made the study visible while it was being created, rather than presenting one long internal form whose result only engineering could predict.
How I got there
I reviewed the existing system, mapped the study setup and worked through the pain points with the employees who used it.
The creation flow was divided into clearer sections and built from reusable components.
What it changed
The dashboard also gave stakeholders clearer access to study status, reducing the need to ask the operational team for every answer.
Studienübersicht › Neue Studie
Study
Sub-studies
Languages
Content
Schlaf und Aufmerksamkeit bei Erwachsenen
Where in the brain is disgust processed?
Wir untersuchen, wie die Schlafdauer die Aufmerksamkeit beeinflusst. Die Messung findet im Schlaflabor Nord statt.
We study how sleep duration affects attention. Measurements take place at Schlaflabor Nord.
2 Termine · je 90 Minuten · Aufwandsentschädigung € 152,00
2 appointments · 90 minutes each · compensation € 152.00
Kein Metall am Körper. Bitte 24 Stunden vor dem Termin keinen Alkohol trinken.
Outcome
The team reported faster work and no remaining workarounds.
According to team feedback, the redesigned dashboard removed the workarounds that had become part of daily use. New colleagues could learn the system with less support as the product and engineering team grew from two to five.
No timed baseline was recorded. The most useful operational measures would be time to create a study, onboarding time, and active studies managed per employee.
Studienübersicht › Schlafstudie › Probandenliste
01.08.2024 14:4022.08.2024 14:32€ 152,00
01.09.2024 08:30€ 250,00
24.04.2024 16:10€ 50,00
22.04.2024 08:50€ 90,00
Screening-ID ändern
Before you can change the patient’s status to Screening, please check the screening ID and update it if necessary.
Datum bearbeitet: 22.08.2024
Are you sure you want to change the patient’s status to Screening?
The change is visible to the study site immediately and is logged.
Team-reported outcome
Faster work, fewer dependencies and no remaining workarounds.
- Workarounds remainingNone
- Team size2 → 5
- Timed baselineNot recorded
02 · Landing page relaunch
Participants and B2B clients needed different paths.
Participants had to understand a study and decide whether to apply. Business clients had to understand Probando’s capabilities and decide whether to make contact.
- 16Languages
- 2Audience-specific journeys
- 1Pre-qualification step before application
Two audiences, two outcomes
Participants needed a clear route to an eligible application.
The intended participant outcome was not raw sign-up volume.
It was completed applications from people who could realistically enrol in a study.
The participant path explained what a study involves, what the time and effort are, and how compensation works. A short pre-qualification step clarified fit before the full application reached operations.
Dein Platz für Studien
Finde klinische Studien, Produkttests und Umfragen.
Observed
Business clients needed evidence of a reliable partner.
The business path focused on capabilities, process and credibility, with a qualified enquiry as its intended outcome.
Matomo, Google Analytics and surveys were used to review behaviour and understand both audiences. No post-launch conversion uplift is claimed because a verified before-and-after result was not available.
Outcome · Proposed measure
I separated the journeys and gave each audience its own structure.
The participant path explained the study, clarified fit, showed compensation and placed a short pre-qualification step before the full application.
The B2B path focused on capabilities, credibility and contact.
Schlaf und Aufmerksamkeit bei Erwachsenen
2 Termine · je 90 Minuten · Aufwandsentschädigung € 152,00
Bewerbung zur Studie
Um an der Studie teilnehmen zu können, benötigt das Studienzentrum von Ihnen noch ein paar persönliche Informationen
Required information
Um an der Studie teilnehmen zu können, benötigt das Studienzentrum von Ihnen noch ein paar persönliche Informationen
The relevant product outcomes were
Qualified applications, completed applications, participants accepted into studies and B2B enquiries.
- Participant measureEligible completed applications and enrolled participants
- B2B measureQualified business enquiries
03 · Design system
The existing library was slowing both redesigns.
Design and engineering were rebuilding the same patterns.
The current design system was difficult to adapt, inconsistent across files and not reliable enough to guide engineering.
I created a shared foundation with Figma variables, tokens, documented states and Storybook references.
- 100+Figma components rebuilt and created
- 3Products using the same design system
- 1Shared source of truth for design and engineering
What was costing the business
Observed
Duplicate components consumed delivery capacity.
The same interface elements existed in several versions across files and products.
Changes had to be repeated, consistency depended on individual review, and every new feature started with decisions the team had already made elsewhere.
This was confirmed through discussions with the CTO and engineering, supported by my own observation across the product work.
The system was built alongside the regular roadmap. No separate pause or system-only phase was required.
Accessibility inside the component
Accessibility was included before the component reached a screen.
Focus, contrast, errors, disabled states and readable structure were defined within the patterns.
This reduced the risk that accessibility depended on a final review or on the person assembling the page.
What I built
The system connected design decisions, component behaviour and engineering documentation.
One foundation supported the internal dashboard, ProbandoPay and the public website.
Figma variables and tokens defined the visual rules. Components included their relevant states, long-content behaviour and accessibility requirements.
Documentation and Storybook references made the intended behaviour visible to engineering.
Multilingual layouts were considered inside the patterns rather than corrected separately in each product.
Outcome: a usable shared foundation, clear documentation and consistent adoption across the active projects.
Wie wirkt sich ausreichender Schlaf auf den Alltag aus?
Untersucht wird, welche Wirkung eine regelmäßige Schlafdauer auf Aufmerksamkeit und Wohlbefinden hat.
- 1color/brand
- 2radius/6
- 3space/12
- 4type/hanken
What changed
Design and engineering
Connected through variables, tokens, documentation and Storybook references.
The dashboard, ProbandoPay and website relaunch reused the same foundation. Engineering could refer to documented behaviour, and the references made the expected implementation clearer.
- color/brand
- radius/6
- space/12
- type/hanken
Schlaf und Aufmerksamkeit bei Erwachsenen
Wir untersuchen, wie die Schlafdauer die Aufmerksamkeit beeinflusst. Die Messung findet im Schlaflabor Nord statt.
Die Teilnahme besteht aus zwei Terminen mit je 90 Minuten und einem kurzen Online-Fragebogen im Anschluss.
2 Termine im Schlaflabor Nord, je 90 Minuten. mehr anzeigen
€ 152,00 nach dem zweiten Termin. mehr anzeigen
Auswertung ausschließlich pseudonymisiert. mehr anzeigen
Nach der Anmeldung meldet sich das Studienzentrum innerhalb von drei Werktagen für die Terminvereinbarung.
Studienübersicht › Probandenliste
Outcome
A system that could be adapted in design and understood in development.
04 · Payment system
One payment process had to work for participants, study sites and clients.
I designed the participant experience, authentication and identity verification, payout status, site operations and client reporting from zero.
- 3User groups: participant, study site and client
- 0 → 1Product design from first rules to complete flows
- Full scopeOnboarding, authentication, identity verification, payout status, operations and reporting
One payout created three different jobs.
Participants needed to verify their identity and understand when they would be paid.
Study sites needed to operate the process.
Clients needed reporting and traceable payment information.
The flow also had to work for people with different levels of digital experience.
The decisions that shaped the product
Identity verification was treated as an important interaction, not as a standard sign-up step.
Requirements, error states and next steps were written in plain language. Accessibility was included in the core flow rather than added after the interface was complete.
Failure states explained what went wrong and what could be tried next.
The user could recover without restarting the whole process.
I defined the product structure and designed the landing page, authentication, identity verification, payout status, study-site workflows and client reporting.
- Payout status answered the next question before support had to.Participants could see the current state, what was missing and what would happen next. The status was designed as part of the product experience, not as internal operational data.
- Reporting connected the payout to its evidence.Study sites and clients could work with the payment record and its supporting information inside the product rather than treat evidence as a separate export.
Studienübersicht › Zentrum › Probandenliste › Termine des Probanden
Überweisung
Please check the patient’s release details.
ProbandendatenÜberweisung
Please check the patient’s release details.
Outcome
Product handover
My deliverable was a complete 0 → 1 product design, from information architecture and user flows to high-fidelity prototypes.
My contribution
The four projects share one decision: repeated work should be solved at the system or product level, not by asking people to work faster.
- Product designDesigned operational, payment and public experiences across three products.
- Design systemsBuilt a shared Figma foundation with more than 100 components, variables, tokens, documentation and Storybook references.
- Product analysisConnected design problems to delivery capacity, operational dependency and application quality.
- AccessibilityIncluded accessible states and WCAG guidance in components, flows and mentoring.
More projects
What else I have built.
TrigonosDecades of geodata, one decision interface.I designed a geomonitoring system from zero for field engineers, analysis teams and clients. It has been in daily use since 2021.
RefinqFrom a pitch deck to a funded product.I turned existing research and founder input into an interactive funding prototype and a documented design foundation.
MaterialSpaceTrust decides whether an AI product gets used.I replaced prompting with structured interaction and validated the approach with 60 participants.