Grubhub - Scheduling
Designed an experience for our Drivers to view and edit their current and upcoming schedule.
The problem
The Plan work is part of a giant effort in Grubhub’s Driver redesign. The goal of this project is to align with GH’s design system, Cookbook. In addition, our Drivers had some incredibly difficult experiences with scheduling. What happens is that when Drivers want to add times they want to work, they see a screen with a bunch of dates and times and they can add times when they want. The problem is that the dates and times are open to all Drivers. This means that thousands of Drivers are competing against each other in order to pick the exact time.
Old designs
Hypothesis
We believe that in redesigning our scheduling experience to include recommending blocks and educating drivers on the best times to work that it will increase core KPI's. Enhancing our scheduling experience with information that is most relevant to our Drivers will increase recommended block pickups, reduce drop blocks, increase drive time, and increase scheduling accuracy.
Identifying the flow
Once we ran user tests on the existing experience, the product manager and I went over our existing data and KPI's. We worked on a product brief together (PRD). To help all our partners understand the problem, we established a basic flow so that we can highlight all the areas around the new experience.
This flow uncovered lots of challenges because we were changing the way our system works. Automatic block selection is a new concept and comes with a lot of new logistics patterns that were important for our team to discuss.
Concept testing
Once agreement was made with product and engineering partners, I then immediately started testing different concepts of how we can help solve for this problem. I was playing around with a bunch of different UI ideas around scheduling. Because this a project that depended on the preferred availability work, it was imperative to take those decisions and learn about the impacts on the schedule itself.
User testing concepts
Conclusions
Usually after initial rounds of concepts, we would throw ideas on usertesting.com or work with UXR to get in person testing.
Learnings
The calendar view made Drivers feel that they can plan for their future.
Adding recommended blocks was something Drivers immediately felt they would add.
There were questions around recommended blocks and if they would actually make more money for our Drivers.
Drivers wanted to see all different order types such as catering orders so they can plan their week accordingly.
Lots of distrust with Grubhub in general
Final designs
Through our learning, we chose to move forward with a calendar view approach. This approach was intended to help our Drivers thinking more long term. Seeing your full week ahead and being able to switch week to week would help in accomplishing that goal. In addition, showing recommended blocks within the calendar view would ideally increase the block pickups and maximize the earnings for our Drivers