Start a project
All work

Case study · Product role

Designing for the non‑specialist.

VisualDx was built by a dermatologist, but many of the clinicians who open it aren’t dermatologists. Across three shipped features I kept working on the same problem: making a specialist’s tool usable for nurse practitioners, primary care physicians and the generalists who see skin every day.

VisualDx logo beside three shipped features: a Skin Lesion Type picker with photo cards, a Compare Diagnoses screen, and a Saved list with a sort menu
Client
VisualDx, clinical decision support for physicians and medical teams
Years
2023 – 2024
My role
Product Designer, then Product Manager: flows, iterations, specs and handoff
Team
VisualDx product and engineering
Scope
Three features across the web app and the mobile app
Status
All three shipped. In VisualDx, at visualdx.com
The problem
The app assumed a specialist’s vocabulary and habits, and the clinicians who most needed help didn’t have them.
What I did
Designed three features, each around one moment where a non-specialist got stuck, worked each through several versions, and tested two of them with 21 participants.
Where it landed
Lesion type examples on the web, and CompareView plus Saved Items and Recent History on mobile. All shipped.

Three moments a generalist gets stuck

  1. Naming what you see on the skin, in dermatology’s terms.
  2. Telling look-alike diagnoses apart.
  3. Getting back to a case you were already working on.

Who this was for

Our Principal Design Engineer described the app as a pilot’s console built for super users. The clinicians in my home screen study were the opposite: nurse practitioners and physicians who see skin problems every day without being dermatologists. They know what they’re looking at clinically. What slowed them down was the tool assuming a specialist’s words and workflow.

How VisualDx works

A clinician describes a patient in findings: a lesion type, where it is, symptoms, age and sex. VisualDx returns a ranked differential, grouped by type, such as infectious or drug-induced, with photos across skin types. Each diagnosis page then explains the condition and lists the look-alikes to rule out. Every one of those steps assumed the user already spoke dermatology.

My role and how the work moved

  • Two roles. I was the Product Designer when Saved Items and Recent History began in 2023, and the Product Manager by the time lesion examples shipped in 2024. In both I owned the flows and the files.
  • Scoped as tickets. Each feature lived as a numbered ticket (UD-153, UD-221, UD-227) with its own Figma file.
  • Versioned in the open. Every file keeps its iterations, a discard area for directions I set aside, and dated wireframe rounds.
  • Specs beside the screens. I wrote behavior notes under the wireframes, covering defaults, time windows, confirmations and empty states, so engineering had rules, not just pictures.

1 · Describe

Lesion type examples

Building a differential starts with describing the skin finding: macule or papule, plaque or nodule. Dermatologists know those terms by heart; many other clinicians don’t. On the web, that step was also buried in a crowded home page.

The original VisualDx web home with a COVID-19 resources sidebar, a grid of eleven problem areas, and a row of links, with problem areas outlined in red
The web home before. My red outlines mark what was crowding out the main task.

An early direction asked for the specific lesion type right away, from one grid of nine look-alike types, with a “Not sure? Include all lesions” link as the escape hatch. I set it aside. The shipped flow splits the choice in two: pick a broad category by photo first, then the specific type, with a plain definition and examples one tap away.

Discarded direction: Dermatology All Skin Types, a three-by-three grid of specific lesion types with View Examples, Continue, and a Not sure link
Set asideNine specific types at once, before seeing a single photo
Shipped: Skin Lesion Type picker with photo cards for eight broad categories
ShippedEight broad categories, each a photo with a scale bar
Redesigned web home: Clinical Tools for Look up a diagnosis, Build a differential and Start with a photo, and My Stuff for History and Saved Skin Lesion Types: cards for smooth papule, smooth plaque, cyst, smooth nodule, hemorrhagic papule and sclerotic skin lesion, each with a definition and View Examples
Top: the new web home, with clinical tools on one side and your own history and saved items on the other. Bottom: after the photo step, each specific type carries a plain definition and a link to examples. The gray boxes are image slots.
Web homeBuild adifferentialPick a categoryPick a lesionby photoChoose thespecific typeDifferentialView examples,then returnNot sure?Web homeBuild adifferentialPick a categoryPick a lesionby photoChoose thespecific typeDifferentialView examples,then returnNot sure?
The shipped lesion flow. Examples sit one step to the side, so checking a term never sends you back to the start.

The decisions

DecisionWhat it replacedWhy it helps a generalist
Two steps instead of oneOne grid of nine specific typesStarts with an easy, visual choice and narrows from there
A photo on every categoryGray tiles and term-only labelsLets you match what you see before you can name it
A scale bar on each photoNo sense of sizeSize separates look-alikes, like a papule from a nodule
The term stays, with a definitionThe term aloneTeaches the vocabulary instead of hiding it
Examples one tap from each typeNo way to checkA clinician can confirm a choice without leaving the flow
Four categories to startEleven problem areas on the home pageFewer, clearer doors into the same tool

What I’d measure

  • Completion. How often a differential started from the web is finished, before and after.
  • Backtracking. How often people go back from the specific-type step to change their category.
  • Examples use. How often View Examples is opened, and whether those sessions finish.

2 · Decide

CompareView

Every diagnosis page lists its differential: the conditions it’s easily confused with. In the app that list was plain links, so comparing two look-alikes meant opening one, memorizing it, going back and opening the other.

I explored a card deck you swipe through, keeping or removing each diagnosis from a working set. The shipped version kept that card anatomy (photo, key symptoms, prevalence, who it affects) but replaced swiping with selecting from the list clinicians were already reading. The current diagnosis stays locked in, so there’s always something to compare against.

Explored concept: a swipeable deck of diagnosis cards with match percentages, and Remove and Add to Working Set buttons
ExploredA swipe deck of diagnoses
Psoriasis page with a Start Comparison button above the differential list
StartFrom the differential
The differential turned into checkbox chips, with psoriasis locked in the tray
SelectThe current diagnosis stays in
Five diagnoses checked in the tray, with a Compare button
Five pickedReady to compare
Compare Diagnoses: stacked cards with a photo, key symptoms, prevalence and demographics for each diagnosis
CompareOne card per diagnosis
The compare view after removing one diagnosis, with the remaining cards closing up
Remove oneThe rest close up
The first screen is the explored concept; the rest are the shipped flow.
Diagnosis pageStartComparisonSelect fromthe differentialCompareCards sideby sideDecideRemove one;cards close upToo many?Diagnosis pageStartComparisonSelect fromthe differentialCompareCards sideby sideDecideRemove one;cards close upToo many?
The shipped compare flow. Removing a card reflows the rest, so you can narrow down without starting over.

The decisions

DecisionWhat it replacedWhy it helps a generalist
Start from the differentialNo way to compareThe need to compare shows up while reading that list, so the entry point is there
The current diagnosis is locked inPicking from scratchThere’s always a reference point to compare against
A tray that counts your picksSelections you have to rememberYou can see what you’re comparing before you commit
The same card anatomy for every diagnosisSeparate full pagesPhoto, symptoms, prevalence and demographics line up for scanning
Remove and reflowStarting overNarrowing down is part of the flow, not a reset

Looking back at the swipe deck

My files don’t record why the deck was set aside, so this is my read today, not the reason at the time. A deck shows one diagnosis at a time, and comparing needs two or more in view. Swiping also asks for a keep-or-remove verdict on each card, which is the judgment a generalist is opening the tool to get help with. Selecting from the list the clinician is already reading avoids both.

What I’d measure

  • Use. How many diagnosis sessions start a comparison, and how many diagnoses people compare.
  • Narrowing. How often people remove cards before leaving, a sign the view is helping them decide.
  • Next step. Whether a comparison is followed by opening one of the diagnoses, versus abandoning.

Testing the new flows

I tested the redesigned lesion and compare flows in an unmoderated Hubble study, setting the new designs against the flows they replaced.

21participants
2flows tested: lesion types and CompareView
Mostpreferred the redesign to the previous flows

Both then shipped. The measures above are what I’d use to check that the preference holds up in daily use.

3 · Return

Saved Items and Recent History

Clinicians come back to the same diagnoses, often mid-case. I designed where that history lives and wrote the rules for how it behaves alongside the screens:

First searchSearch tips, with nothing to show yet
Returning within a monthRecently Viewed under the search bar, grouped by date
Gone about two monthsHistory has aged out, so search resets to the tips view
Saving a diagnosisThe outlined bookmark fills, and a toast confirms it
Nothing saved yetAn empty state that says what will appear here
Items savedA list, sorted by most recently saved
SortingOne choice at a time: Recently Saved, Oldest First, A to Z, Z to A
Cleaning upSwipe to delete, then confirm
Every state from my notes. Each one tells the clinician what is happening and what to do next.

Sorting is where my notes record a real decision. The first version used checkboxes, which implied you could sort alphabetically and by date at the same time. I wrote:

“I felt it was best to utilize dropdown menus in order to mitigate toggle confusion.”

My design notes, October 2023

The final version went one step further: a single-choice menu of Recently Saved, Oldest First, A to Z and Z to A, where only one option can ever be on.

Search opened on a Recently Viewed list grouped by date
Recently viewedRight under search
Saved tab empty state: Nothing here, yet! Your saved diagnoses will show up here.
Empty stateSays what goes here
Earlier sort design: a bottom menu with checkboxes for Alphabetical, Date Added and Last Viewed
Sort, beforeCheckboxes invited two sorts at once
Final sort design: a single-choice menu with A to Z checked
Sort, afterOne choice at a time
Saved list with one item swiped to reveal a red delete button
DeleteSwipe, then confirm
The shipped app: search with Recent Activity and Saved Items tabs, and recent diagnoses grouped by date
In the appShipped
The first five are my wireframes from September and October 2023; the last is the shipped app.

What I’d measure

  • Return speed. Time from opening search to reopening a diagnosis viewed before.
  • Saving. How many active users save at least one diagnosis, and how often they open Saved.
  • Clutter. How often items are deleted, a sign the list stays useful rather than piling up.

Principles that carried across

Recognize before you recallShow the photo, the list or the recent item, so nobody has to produce the answer from memory first.
Keep the expert term, add the plain oneEvery term stays, with a definition beside it. Generalists learn the language instead of being shielded from it.
Never strand the userEscape hatches, empty states that explain themselves, confirmations, and a way back from every step.
Write the rules, not just the screensTime windows, defaults and edge cases go in the file, so what ships behaves the way it was designed.

What I learned

In a tool built by experts, the expertise can become the barrier. Each of these features moved one specialist assumption out of the way: the vocabulary, the memory needed to compare, and the effort of finding your place again. None of them hid the precise terms dermatology depends on. They just stopped requiring you to know them before you could start.

A note on sources: this case study is rebuilt from my original Figma files, including my design notes, iterations and discarded directions. Where the files don’t record a reason, I say so instead of filling one in.

Credits

Design and product
Jaylon Ballard
Team
VisualDx product and engineering
Note
VisualDx is a trademark of VisualDx, used here only to name the client. Clinical images © VisualDx.