H1 People Search
Business Need:
Search is the top activity users of H1’s HCPU platform perform, but it is tedious to operate and overwhelming to users, not meeting their needs.
Goal
Redesign H1’s people search so that it is more intuitive and knowledgeable of MSL’s people discovery goals.
Timeline
1 Year, 2021-2022
My Role
Part-time focus for over a year
All UX/UI design
Conducted new user research (stakeholder interviews, user interviews, contextual inquiry, zendesk requests, full-story observations, mixpanel metrics)
Worked closely with PM(s)
Created new components for existing design system
Work with Engineering on from discovery to implementation
Background
Medical Science Liasons
MSLs are employed by pharma companies to find influential doctors so they can exchange scientific knowledge in order to advance their business goals.
The Medical Science Liaison is a role within pharma companies.
They are highly educated, scientifically trained field staff that are able to build rapport with Key Opinion Leaders (KOLs) in various therapeutic areas of research.
MSLs find and engage with doctors who are KOLs to educate them on ongoing research, recruit them to run clinical trials, or collect insights from them on new drugs.
Therefore them must identify and discover KOLs by searching through H1’s database of doctor profiles.
Research
8 stakeholder interviews
2 new user interviews/contextual inquiry
Affinity mapping ~185 zendesk requests
Hours of fullstory observations
mixpanel metrics
Research Insights
User are asking for easier to use, smarter, more flexible search to yield relevant recommendations with clear feedback.
Users often asked for
Ability to exclude criteria
Various specific new filters
Better usability around current components
More ways to sort and display the results
Visualizations
From observing users’s behaviors via mixpanel and fullstory
Users most often used the global search bar to access a particular profile by name
Used certain filters more than others
Looking beyond the surface asks
Users have a hard time finding KOLs because –
H1 makes a “database search” available to the user, with no insight into what they are trying to accomplish
Examples:
min/max filters → bucketed segments
MSLs have territories
In addition to visual design and best practice improvements, we should provide easy presets that makes discovery less tedious.
Additionally, make allowances for the behaviors we understand from users
let users navigate to a profile easily because this behavior is more common than real “searches”
surface the filters users use most often (location, tags)
Create an easy to use search interface that can be made iteratively smarter, tailored to our users existing workflows, so that MSLs can discover and identify HCPs that they should be engaging with.
Design Goal
Iterations and Sketches
More Projects
-
DFCI Curation Platform
-
Broad's COVID-19 Dashboard
-
Terra Data Explorer