Led redesign of a core sales funnel connecting physicians with brand representatives, using stakeholder interviews and data analysis to address drop off and clarity issues. Simplified the flow through progressive disclosure and clearer entry points, resulting in a more focused, conversion oriented experience.
Role
Timeline
Area
Lead Designer
2 months
Feature enhancement
Background
Used in segmentation, data integration, routing etc
Brand awareness through to conversion
Featured on majority of US pharma sites
How it started
Low conversion rates prompted a direct challenge from a brand lead: can we do better? Understanding their frustration, I began a fact-finding effort to identify the underlying causes rather than jumping into solutions.
Analysis
Previously, the forms capabilities and performance were not questioned. For years, it was seen as just a standard form despite high importance. Outdated design and mental models began to catchup.
Balancing trade-offs
User analytics were not available :(
User events were not implemented at the time aside from conversation rate and time-on-task. This made it difficult to understand user habits and benchmark performance.
We couldn't speak to the user :(
Privacy and time constraints limited access to rare-disease physicians meaning I had to rely on alternative research methods.
Research
Patients were finding ways of submitting the form. This is something we'd to prevent as it could mean sales reps may be exposed to sensitive medical information creating compliance issues.
Limitations due to previous implementation meant that the feature lacked unique customisation for brands often causing frustration.
Once we knew the problem, a brief was formed
Redesign the lead gen form to increase leads and prevent patient submissions
Ideation
A design sprint surfaced several approaches. We prioritised a concept using progressive disclosure and representative imagery to address trust and cognitive load.
Testing
After finalising a concept, we moved to testing. Initially, we thought it would be useful for each physician to be assigned a rep based on their ZIP code. However, testing indicated that this was causing confusion.
Being assigned a rep caused confusion for physicians who may already have been assigned a rep or assigned incorrectly.
Form confirmation now features multiple rep images and removes specific rep names to keep assignment open.
Two rounds of testing with separate groups confirmed the approach and surfaced positive feedback.
% preferred new form
% more trusting
% more responsive
% more engaging
Handover
Knowing how the final concept will look, I collaborated with cross functional partners to define KPI's and phase out the project into a priority based roadmap.
Redesign a feature for increased conversions connecting reps to physicians
UX overhaul to simplify and streamline the user flow to prevent exit rates
Introduce relevant trust signals to improve user expectations
Opportunity to remove redundant sections
Allow for easy navigation and editing between steps
Introduce motion design to bring delight and polish
Brand-specific steps for unique needs
Ongoing collaboration with engineering aligned expectations early, reduced implementation risk, and supported a clean handover into build.
Updates on current projects and backlog - sets expectations
Components
Specifications
Examples/flows
Audits of solution determines ‘done’ status. Revisions made.
Solution
Leading with the physician’s request clarifies intent, reduces friction, and makes the experience feel immediately purposeful.
Solution
Solution
Solution
Impact
Detailed event tracking initiated
This redesign focused on improving the experience and reducing risk, not just increasing form submissions.
Reducing cognitive load and clarifying expectations shortened time to complete the form increasing potential leads
Removing patient pathways and reinforcing professional-only messaging reduced the likelihood sensitive data submission
Introducing trust signals increased credibility supporting completion
Reflections
Conversion problems are trust problems
I believe drop-off was driven more by lack of reassurance than form length. Introducing trust signals earlier had more impact than removing fields.
Data capture needs to be a priority
Implementing step-level events shifts conversations from opinion to evidence and will enable more confident iteration in the future.