Workday Scheduling Mobile
Workday
180+
store implementations (at launch)
20K+
users (at launch)
Role: Mobile Product Designer
Team: 1 Web Product Designer, 1 Researcher, 3 Product Managers, 12 Engineers
Empowering frontline workers to have a say in how, when, and where they want to work.
Businesses are struggling to retain frontline workers due to inconsistent schedules, low employee engagement, and lack of career growth.
In 2019, my team was founded and tasked to find a way to give these workers more control over their work. In October 2021, we launched Workday Scheduling, a new AI-powered solution that matches business labor demands with worker preferences.
Due to NDAs, I cannot publicly share all details of my work. The information below is my own and does not necessarily reflect the views of Workday.
Introduction
Our "stealth startup" team was founded with a primary objective to build a smart workforce scheduling solution
For years, Workday customers have been asking for a scheduling solution that was fully integrated with our Workforce Management suite so that they could plan schedules without using disjointed third-party applications.
HCM
Time
Tracking
Absence
Management
Payroll
Scheduling
Scheduling is the final piece of the Workforce Management suite
With this in mind, our team was created. We were required to work in secret, aiming for the ideal solution not limited by existing Workday constraints or design patterns.
We kicked off this project by gathering 12 customers into a design partner group (DPG). The goal of this focus group was to collaborate, exchange ideas, and provide us real-time feedback. Many of them would eventually become our pilot customers and early adopters.
Research
To better understand the frontline workforce, the research team spent 4 months interviewing 70+ workers and managers around the world
In speaking with workers, we discovered several insights including:
Some workers don't own a computer, phone, or car; many are not tech-savvy
Inconsistent schedules impact workers' financial, mental, and physical health
Getting a shift covered is stressful; the worker gets penalized if coverage isn't found
Workers are frustrated that they have little to no control over their careers
A mapping exercise of shift coverage transactions between workers, teammates, and managers
Brainstorming & concept design
Leveraging our research, we solidified a high-level solution to give workers more decision-making power in the scheduling process
Before a schedule is created: Workers can set their preferences such as preferred role, department, and work times. Workday's AI would then generate a work schedule that balances business needs with worker preferences.
After a schedule is created: If workers can't work a shift, they can give or swap their shift with a teammate. They can also post it to an open shift board for others to pick up or drop the shift in extreme cases. The Workday AI will validate the transaction based on specific requirements, reducing manager oversight and ensuring the shift gets covered quickly.
Some early explorations into scheduling UIs, preference and availabilities, and AI chatbots
Validation
Positive usability test results validated our design direction, leading the way to our proof of concept
My concepts were regularly shared with the DPG, allowing us to continually refine the experience. As the designs began to mature, we conducted usability tests with the DPG's frontline workers to measure task success rates and task completion times.
The tests performed well with minor micro-interaction and feature discoverability issues. However, these were easily addressable by adjusting our components and configurations.
In-person usability test with a frontline worker
Hitting a roadblock
Challenges integrating our bespoke experience with the Workday mobile app forced a redesign
After making usability improvements, our proof of concept was successfully greenlit by leadership. Our "stealth startup" graduated into a full product team so we could begin integrating with the official Workday app.
This, however, turned out to be the most difficult phase of our project. Our custom app was not compatible with the mobile tech stack. After failed negotiations, we were required to make major compromises including a redesign using the existing mobile framework.
We documented the consequences of the newly compromised designs
The employee mobile app MVP
Finalizing the employee experience
Over the next few months, I paired closely with the design systems team to redesign the app. I then collaborated with the PMs, engineers, and accessibility advisors to conduct bug bashes and make final adjustments.
Helping workers stay compliant via their schedule and shift details
Giving workers more control and flexibility via work preferences and shift transactions
Keeping workers up-to-date with a smarter check/in out integration (Check In/Out Designs: Ivy Lam)
The manager mobile experience
With the employee MVP completed, I was tasked to design a mobile app for managers to manage schedules on-the-go
At this point, my design partner and team had already built out majority of the desktop experience, which is broken into two sub-products.
Labor Optimization: historical data is added to help the AI forecast labor demand.
Labor optimization (Design: Jared Hirata)
Scheduling: labor forecasts, employee preferences, labor laws, and more are fed into the AI/ML algorithm to generate schedules. The AI reduces time spent planning schedules from an average of 4 days to minutes, saving managers a significant amount of time.
Scheduling (Design: Jared Hirata)
Approach & concept design
Unwilling to run into the same roadblock, we presented data from the compromised designs to win leadership support, allowing us to build our ideal experience
Our researcher conducted simulated tests to compare task completion times between the original and compromised experience. One primary task which originally averaged 1-2 minutes was now taking 4-5 minutes to complete. The massive difference alarmed our stakeholders, encouraging everyone to work together to lift the blockers.
I involved the mobile and design system teams to ensure our alignment. Given the density of our scheduling solution on desktop, I worked to make the schedule was scannable in small devices by modernizing the mobile UI and introducing new card components.
Early wireframes of the manager mobile app
The manager mobile MVP
Finalizing the manager experience
With leadership backing, we were able to move from design to production in less than 6 months. The current iteration of the manager mobile app serves as the on-the-go solution for keeping track of and making adjustments to the live schedule.
Helping managers stay on top of their workforce
The Launch
Workday Scheduling officially launched in October 2021
After two and a half years of hard work, we launched Workday Scheduling at Workday's Conversations For A Changing World conference.
At launch, our DPG early adopters who had been piloting our product across 180+ stores, totaling 20,000+ workers and managers signed on to be our first customers.
Reflection
I am incredibly grateful to have been part of such an ambitious project and talented team. It wasn't easy diving into complex topics like labor laws or working through the COVID pandemic. However, our team managed to overcome these challenges and build Workday's first AI-powered product from scratch.
While I was no longer at Workday to observe and measure the success of Scheduling, conversations with our pilot customers indicated we had already begun empowering our frontline workers and managers alike. I've also taken along with me a few learnings:
"Starting with a skateboard" — the skateboard model drives the way our team tackles product development. Sure, our MVP may not be perfect right away — but it's enough to get customers started and enough for us to begin learning how to adjust course.
Driving alignment with data — data is so important when communicating with stakeholders, especially after sharing how extreme the task completion times varied between our original and compromised experiences.