top of page
lOTUS (3).png

Enhancing clinic-wide scheduling and transportation

Solving the unique model of our care organization by creating a scheduling tool that balances patient preferences, transporation needs, center oversight and mistake prevention.

The Challenge 🕰️

When your scheduling an appointment for the future feels like the past


At Oak Street Health, provider time was gold—but our scheduling tools were stuck in the stone age:

  • Basic tools that barely scratched the surface of patient needs

  • Welcome Coordinators juggling multiple systems

  • Transportation coordination that resembled air traffic control without radar

  • Schedule visibility about as clear as a foggy window

  • Patient preferences buried in notes like treasure without a map


The result? Wasted provider time, frustrated staff, and patients stuck in scheduling limbo.




My Role 🎨

Choreographing the scheduling dance as Design Lead


I stepped in as the scheduling maestro:

  • Led a design trio (myself, one UX designer, one UI designer)

  • Partnered with 6 engineers who turned our vision into reality

  • Owned research initiatives that uncovered the real scheduling pain points

  • Played diplomat between stakeholders with competing priorities

  • Held ground when "move fast" threatened to break essential things



End Users 👥

The Heroes Behind the Screens

Welcome Coordinators (WCs)

  • Scheduling Time Sink: Spent 5-7 minutes per patient booking appointments, severely limiting how many patients they could serve each day

  • System-Switching Nightmare: Constantly jumped between different systems for patient notes, preferences, scheduling, and transportation coordination

  • Transportation Chaos: Managed complex patient transportation using manual spreadsheets and endless phone calls to drivers

  • Invisible Provider Availability: Couldn't see multiple provider schedules at once, making it impossible to quickly spot available slots

  • Lost Patient Preferences: Critical patient needs (language, provider gender, time preferences) were buried in chart notes rather than built into the scheduling process





While welcome coordinators were the main user group we also considered Call Center Service Agents, Transitions care team members, Transportation leads and Practice Managers.


Methodology 🧭


Human-Centered Design Process
ree

Starting with a kickoff workshop
  • Limited design sprint workshop over several days to determine shared vision

  • Ideated with 22 diverse users and executives to gain alignment

  • Everything done remotely



ree

Honing in on the Vision 👁️ 


Build or Buy Decision

We used the original workshop to understand if scheduling was something that had to be built vs something we would make in house. The workshop was crucial in final alignment.

ree

Iterations

Figuring out the flow was difficult in the first few versions. The first few iterations had too many steps at once. We also tried scheduling formats of our competitors but did not like how unclear it was to select time slots. We needed to do something different.



Sharing the Vision

By narrowing down the features, we were able to


ree


Key Design Breakthroughs 🚀


1. Patient Preference Powerhouse

Because patients aren't one-size-fits-all, and neither should their scheduling be


The Problem: Patient preferences were floating around in chart notes like unanchored ships, never making it into the actual scheduling workflow.


The Aha Moments:

  • Discovered some patients needed scheduling Tetris (urgent appointments + unavailable providers)

  • Learned patients have scheduling rhythms—some are morning birds, others afternoon folks

  • Found care teams naturally thought in AM/PM blocks, not hour-by-hour slots


The Solution:

  • Created a preference panel that remembered what patients needed

  • Treated location as a suggestion, not a rule (because patients roam!)

  • Built time preferences that matched how people actually think about their day

  • Made matching patients to providers feel less like blind dating and more like matchmaking

ree

2. Transportation Coordination Magic

Turning the transportation tangle into a well-oiled machine


The Problem: Transportation scheduling was a spreadsheet nightmare that would make Excel experts weep.


The Discoveries:

  • Some patients needed rides every time, others just occasionally

  • Welcome Coordinators were spending more time as transportation dispatchers than schedulers

  • Nobody could see the full patient journey from pick-up to "the doctor will see you now"


The Solution:

  • Created a unified view showing provider time and transportation side-by-side

  • Built in transportation needs right into appointment details

  • Added conflict alerts before scheduling disasters could happen

  • Designed pick-up/drop-off visuals that made sense to actual humans


ree

3. Clinic-Wide Scheduling Reimagined

Breaking free from the tyranny of traditional calendars


The Problem: No one could see the forest (clinic schedule) for the trees (individual appointments).


The Revelations:

  • Providers and vans were the twin pillars holding up the clinic day

  • Schedulers constantly compared schedules like they were solving complex puzzles

  • Care teams cared about both filling slots and keeping providers sane

  • The context around each appointment was crucial for smart scheduling


The Game-Changer:

  • Flipped the script with a horizontal view showing multiple providers at once

  • Added 3-day and 5-day toggle views to spot gaps at a glance

  • Focused transportation displays on what actually mattered: pick-ups and drop-offs

  • Created a scheduling command center that matched how schedulers actually think

  • Made schedule patterns pop visually so problems couldn't hide

ree

Leadership Story 🕵️

Standing up for users when "faster" threatened to derail "better"


The MVP Showdown: When product wanted to ship without the clinic view (calling it an "MVP"), I channeled my inner advocate:

"Without the clinic view, we're asking users to bounce between systems like pinballs. That's not minimal—it's incomplete."


What Actually Happened:

  • Initial reception: "This looks great!" User adoption: crickets

  • Staff quietly slipped back to trusty old Greenway for the clinic view they needed

  • The experience became a powerful case study in what "viable" really means

  • This battle scar informed our approach to future projects, saving us from repeat mistakes


The Impact 📈

When design delivers more than just pretty screens


The Business Win:

  • Launched in less than 5 months (while vendor solutions were still filling out paperwork)

  • Saved millions in subscription fees that would have made vendors very happy

  • Built exactly what we needed, not what vendors thought we needed

  • Beat out fancy third-party solutions that promised the moon but delivered cheese


The User Victory:

  • Scheduling time plummeted from 5-7 minutes per patient to under 1 minute

  • Schedule visibility went from squinting to crystal clear

  • Transportation coordination transformed from nightmare to manageable

  • Patients got appointments that actually matched their needs and preferences



Lessons & Future Vision 🧠


What We Learned:

  • In healthcare, context isn't just king—it's the whole royal family

  • Fighting for essential functionality pays off in the long run

  • Sometimes the best solution is the one you build yourself

  • Always, always design for how users actually think, not how you think they should think


What's Next:

  • Tools for managing holidays and provider time-off that don't require a PhD

  • Seamless integration with out-of-office settings

  • A dashboard showing slot utilization that even executives can understand

  • Self-scheduling that doesn't make patients want to throw their devices


This project didn't just transform scheduling—it proved that deep user understanding plus the courage to advocate for what matters can create something truly game-changing in healthcare. No generic scheduling platform could have delivered this because they don't understand the complexity of healthcare scheduling. We do. And we built it.



bottom of page