Up until recently, a print directory had been used.
When they launched the first searchable online directory, it contained a number of suboptimal UX features..
In v1.0 of the online directory:
Search filters were exclusive; only one could be applied at a time
Arcane medical jargon was used for SEO/keyword searches, rather than common search terms
Map display of results was static, not interactive
Results were exclusive to the entity being searched; linked network providers were not included
To assist engineers in improving search results in this pre-AI project, I generated a synonym list they programed to translate common terms for ailments, symptoms and body parts into the medical specialties that treat them.
This list was initially built in Google docs, and later integrated into the backend of directory search logic.
Our initial feedback session evolved into a series of iterative reviews of new features.
My design review surfaced a number of UX improvements that were implemented:
Keyword tagging rather than filters, so multiple selections could be applied together.
A dropdown list of specialties, to assist patients who don't know medical terms to search
A clinic selector to enhance specificity within the geographic search
Interactive mapped results, with hover-over clinic details and click-through capability to providers
A schedule availability selector, allowing patients to focus on times they can make--both in the search results and provider landing page
A language dropdown to help patients with communication barriers
The directory would serve three integrated groups--all commonly owned, but operating as separate brands, each with a unique website.. Surfacing doctors from all brands across the region was needed to keep referral traffic internal.
Prior legal guidance had advised that doctors could only show up on the website of their legal employer. The concern was that featuring doctors employed by separate legal groups may be construed as a Stark Law violation (by providing material benefit to some doctors). The resulting data separation led to poor UX, negative business impacts, and the ongoing false impression that our internal brands were competitors.
After reviewing the law and consulting decisionmakers, I convened meetings with leadership, legal, and marketing teams to discuss the interpretation and impacts.
As a result, the legal guidance was reversed, opening the gates to doctors from multiple brands appearing together. This allowed patients to see the full slate of their options and strength of our network on any directory they visited.
The current build used across seven states includes a number of design upgrades 1 suggested & worked on:
Search field matches common synonyms for latin medical terms
Insurance filter surfaces in-network providers; Dropdown specialty prompts doctor fields
Filters now function as keywords with multiple simultaneous selections
Availability (video vists, accepting new patients) now appear on seach result cards, no clicks required
Other locations a provider sees patients can be opened
Mapped results show other providers at same location
80% increase in click-through rates and conversions
40% increase in scheduled appointments.
Our nine-clinic system was an unintegrated part of a seven-state healthcare conglomerate. The directory work for us started as a routine implementation of their broad directory product on our standalone website, before the branding transition.
Engineers commented that our feedback was the most detailed and in-depth they had received.
Due to the superior design features of our local directory instance, our version was adopted as a template for the entire clinic system representing thousands of clinics, 34k physicians and 28m patient visits per year.