UX Design Case Study

UberEats
KnowMe

Concept design
My Role
UX Research & UI Design
Approach
Brainstorm Idea
Understand Problem
Competitive Analysis
User Surveys & Interviews
Social Listening
Research Findings
Iterate & Re-evaluate
Feature Ideation
Wireframe Sketching
UI Design
High Fidelity Prototype
A/B Testing
Problem Statement
Being unable to prevent delivery related issues from repeating, frustrates Uber Eats users.
Acknowledged diverse needs of the users in the final feature:
Preliminary Research | Existing Users
Through social listening, I found positive & negative UberEats reviews from the App Store/ Play Store & Reddit.
Trying to understand and find if there were any gaps in the current experiences of UberEats users, led me to look at online reviews where real customers were sharing their own experiences.
‍
I looked at the reviews from the past month, finding the app to have a 4.7/5 star rating. However, over 50% of the most recent reviews were less than 2 stars.

Then I dug deeper to find out why.
Food with bite marks | Bad handling | Ignoring instructions
One redditor left a testimonial about his wife who is wheelchair bound & unable to cook. Her delivery instructions were ignored 3 times by the same driver.
Currently unable to indicate preferences for a driver
(i.e. “I would prefer to not have a delivery person on a bike”, or “I would prefer a driver who is familiar with my neighbourhood/ area”... etc.)
Preliminary Research | Current Drivers
To gain perspective, I explored the POV of current UberEats delivery persons
There is a disconnect between the experiences of drivers & customers. One delivery driver wrote "It's like the customers don't even know there are more orders before theirs".

This really helped gain better perspective on some of the pains and issues.

Some key takeaways included:
‍- Communicating with customers helps prevent customer frustration
- Drivers see tips beforehand & may cancel due to low tips - may be why users wait goes up
- Customers can just click thumbs down without justification
- Not being able to read house numbers at night can also cause delays
Competitive Analysis
Explored the end to end processes of the top 3 food delivery apps in Canada.
All have grocery, delivery scheduling, favourites & ability to send orders as gifts. Customers expressed unhappiness with deliveries.

UberEats offers priority deliveries,  multiple carts.
‍
UberEats does not let customers explain why they gave a downvote for a delivery.
User Surveys
I honed in on experiences with food delivery services, drivers & feedback habits.
Qualtrics Experience Management was utilised to create the survey. It was conducted over the span of 72 hrs, answered by 12 participants, all based in Canada. Majority of participants had used UberEats, and all had tried at least one food delivery app.
Showcasing the experiences by food delivery app users:
99% of survey participants pinpointed at least 1 thing they appreciated about delivery drivers
100% experienced at least 1 delivery issue, with 75% had issues with delays due to drivers
Carefully considered how the target users had vastly different lives & experiences when designing the feature
Target Users/ Persona
People who RELY on the app & NEED their specific preferences to be met
People with experience health issues (i.e. immunocompromised, bed-ridden, physical limitations, social anxiety, depression, & others), that restrict them from going out, making or cooking their meals. They may have limited mobility, can’t go get groceries to cook due to lack of transport due to no car, harsh weather conditions, feasibility, health.
People who use the app on occasions, & have other means of getting food
Might be students, families, professionals who are short on time or want to order in for a change. They may have limited means to pick up food, timely obligations, but have other means to get food (not relying, more convenience)
Ideating
There is currently no way to “avoid being paired with a specific person”; as a driver or customer.
Poor delivery service & bad customers do not have true consequences.
Idea was to give customers the ability to block and or favourite delivery drivers
Asks “would you like it if __ delivered your order again?” giving the user control and allowing them to indicate preference for certain drivers due to good service. If they answer “no” then it would take them to “get help” but instead of just having “past order” and “safety” options, would let them pick to block a driver
Putting control in the users hands without having to go through the process of contacting customer service.
This would incentivize people on better behaviours through more than just a thumbs down for poor service.
Keeping the problem in mind
Experiencing delivery issues & having limited control, causes frustration.
Evaluating the idea
Feasibility/ Viability/ Desirability framework, to assess if this was the best solution.
Key Points:
Feasible?
‘Blocking’ gives control to the user, drivers would need equal feature.
Banned list could lead to more issues, like having less drivers available
Viable?
From a business standpoint, only a small subset of users have had issues that would warrant blocking delivery persons
Desirable?
Based on research, a majority of users are not quick to report minor issues. Negative ones were driver dependent and situational. Favourites feature would be more desired but likely overused.
Needing a New Solution
Needed a more feasible & viable solution that still allowed control for the user.
The “restricting or favouriting drivers” solution may not give the desired outcome (of making people happier with the amount of control they have in their delivery experience) and might cause issues from a drivers standpoint by giving too much flexibility.
Tackling the problem from the account profile.
Ideating New Solution
Implementing a feature of user preferences selections, to allow customers to have a more refined and personalized delivery experience.
Mapped out the user flow of selecting delivery preferences.
Sketched ideas for the UI with notes.
Getting “inside the mind” of the user by letting them choose their top priorities for a delivery.
Really understanding the true preferences - ideally leading to a more custom delivery service experience for the user.
Created the feature; a multi-screen addition to add delivery preferences.
UI Design Prototyping
Used existing UberEats style guide to inform the new feature. Components & style guide created with new icons as well.
Components & style guide created with new icons as well.
Style Guide
Asked participants about their top options for the selections, and if any were confusing.
There were similarities, so I ordered them with these popular ones at the top of the list. Subtext was removed from all options by re-wording the main text to include more specific details if needed.
After the text had been improved, another iteration was made to simplify the flow more..
Version 2
Consolidated everything on 1 screen, reducing clicks & improving usability.
Comments from further testing:
1. The blank space with the 3 gray areas is unnecessary
2. ‘Tell us how you like it’ is redundant/ confusing
Updated:
Multi-screen VS. single screen flow...
A/B Testing
Conducted A/B testing comparing the two prototypes, to understand which worked better.
The hypothesis was that version 2 would preform better in terms of simplicity and having less screens to navigate to.

Also collected general feedback of the overall usability & copy of the feature. With the newly improved versions, shared both prototypes and let participants walk through the screens and flows.

Utilized feedback from 4 participants.
The consensus was that option B was better in terms of a cleaner, visually easier to manage UI.
Final Feature
UI Design of High Fidelity Prototypes
Project Reflections
What I learned
Value of understanding the problem, FIRST.
A presentation on ‘problem first thinking’ helped rethink my strategy, to focusing on understanding the problem before anything else... I was able to hone in on gaps that existed.

Going forward I will create a solid problem statement first as it leads to brainstorming better solutions that truly cater to the needs of the target users.
UX Content Writing
Lack of clarity, feeling overwhelmed by the number of options & the way that things were worded, all played a big part in users perception of an app.

Going forward, I would keep all iterations (to save time if it was necessary to utilize information from past versions).