Define
RESEARCH
Conducting market and user research to identify the problem space.
PROBLEM SPACE
Consumer “Promiscuity” & Churn due to lack of options, deals, fast delivery, or quality is the biggest problem DoorDash is facing, majorly caused by:
Delivery issues
Recurring complaints regarding delivery issues, such as food never being delivered, low food quality, or food inventory issues.
High delivery fees
Customers dislike the high delivery fees from ordering food online, which is why they look for deals on other food delivery apps.
HOW DO I KNOW THIS IS A PROBLEM?
Using forums and social media to gather qualitative data.
Posted the user survey on Reddit, Social Media, and other forum sites to gather qualitative data on likes and dislikes with DoorDash to frame my user interview questions.
From the survey respondents, I selected and spoke to frequent DoorDash users who only use DoorDash and also those who use competitor apps, like Uber Eats and GrubHub, DoorDash’s main competitors. These interviews led to these insights:
-
Ordering in groups: when ordering with multiple people, it is hard for most users to come to an agreement on what to eat and people would prefer to avoid compromising what to eat.
-
High delivery fees: when deciding where to order from, most users refer to deals and cost to decide which app to use.
-
Food quality: most users experienced cold food, delayed delivery times, and food quality issues, such as missed items when ordering from DoorDash.
-
App switching: despite use DoorDash most often, most users switch between apps in search of lower fees, deal,s and restaurant variety.
-
User type: the majority of users do not use DashPass because they don’t order from DoorDash frequently enough to do so and do not want to risk signing up for yet another subscription.
-
Curiosity & Grass is greener: users are naturally curious and even when they know DoorDash offers low delivery fees compared to other apps and a variety of restaurant options, they still are inclined to view other apps for the sake of validating their thoughts.
USER PERSONA
The user persona guided the majority of my design decisions.
This persona was created after user interviews and affinity mapping. It was regularly used throughout the ideation and design process to ensure a user-centric approach. It came in handy during the prioritization of possible features, to find the one that best meet user needs.
FOCUS
How might we provide something of value to infrequent DoorDash users?
After user interviews, I synthesized my notes and added categorized them to uncover insights and arrive at this HMW statement. From there, I brainstormed a few solutions by doing a crazy 8 exercise. These solutions include:
POSSIBLE IDEAS
Sketching possible solutions for high delivery fees and arriving at 3:
Neighborhood order: Neighborhood order with people near you versus only ordering with family & friends, to reduce delivery and location fees and increase customer satisfaction.
Multiple restaurants: Ability to order from multiple restaurants in a single order when ordering with large groups of people, to increase consumer satisfaction.
Offers: Discounts, Promotions, and Incentives, or virtual stamp cards to increase orders per month.
Prioritizing the solutions brainstormed to meet both user and business goals.
USER NEEDS
I made sure to think back to my user persona at this stage of prioritization to identify features with the most impact and reach.
Neighborhood orders with people nearby had the highest impact and reach because most users expressed how much they disliked the high fees from ordering, and most users typically order by themselves.
Impact
Features that address the user goals and pain points uncovered from user interviews.
Reach
Features that would benefit most users by understanding clearly defining which circumstances these features would used in.
COMPETITIVE ADVANTAGE
Referring to competitor analysis to finalize which feature to move forward with.
Business goal: Attempt to reduce customer app switching to increase the company’s monthly revenue, and further product differentiation from competitor apps.
Referred back to two key competitors uncovered in secondary research to make sure I took into account DoorDash’s goal to gaining a competitive advantage. Doing this helped me understand which two key features to focus on:
-
Neighborhood order with other DoorDash app users nearby
-
Ordering from multiple locations when ordering with friends or large groups
BUSINESS GOALS
Bringing everything together finalize the feature to move forward with.
Addressing product differentiation and likeliness of reducing app switching to come to a conclusion on which feature best address these business goals.
Given the main pain point our persona has with DoorDash is the high fees, which is also a main reason why they switch between other apps, it made sense to move forward with the Neighborhood Order feature.
UX SOLUTION
Reducing the delivery fees to increase customer satisfaction and gain a competitive advantage.
I took into account the frustration users feel towards high delivery fees and delivery issues to help reduce app switching.
Although, I’ve come to realize this feature won’t 100% solve the issue with app switching because interviews revealed the curiosity users have about other apps, my hope is to help reduce this problem and help DoorDash gain a competitive advantage.
This feature gives users the option to order from people near them (within ~ a 2-mile radius) versus only ordering with family & friends, to reduce delivery and location fees and increase customer satisfaction.
Design
SKETCHING & USER FLOWS
Two possible task flows for this feature.
Task 1
Join a group order with people within a 2-mile radius of you. Shows restaurants with live orders you can join, but if none of them are what you are looking for, you can...
Task 2
Create a group order with people within a 2 mile radius of you.
DESIGN EXPLORATIONS
Identifying DoorDash patterns I can use in my designs.
Exploring DoorDash patterns to understand how they are used and to reuse them for my designs to ensure consistency.
ITERATIONS
Making changes using crit feedback.
ITERATION 1: VISIBILITY OF SYSTEM STATUS
Focusing on visibility of system status to keep users informed and answer questions they have while moving through the ordering process.
After taking my current designs to group critique for feedback, multiple questions popped up, which helped me narrow down what I needed to revise before moving forward with my wireframes.
-
How much longer will my order take if it’s a group order vs. individual order?
-
How does the delivery time get determined?
-
What is the incentive for someone to create a group order?
-
What is the benefit of a group order?
-
What happens if one person cancels?
-
How much am I saving?
-
What is the maximum number of people per group order?
Revision 1.0
Explored competitor and DoorDash patterns and other mobile design components to use on the new screens and answer, “How much longer will my order take if it’s a group order vs. individual order?”
Revision 1.1
Adding additional information about how the group order works to answer the users questions.
USABILITY TESTING
How effective were these iterations in solving common questions from a new user perspective?
After making iterations from the feedback gathered in design crit, I conducted 6 usability tests which led to two main discoveries.
Concerns with delivery time
Participants questioned how much longer delivery time would take when you set a deadline for the length of time the group order would remain open, and the time it would take for the Dasher to get to you.
No incentive to create your own group order
Participants were hesitant about creating their own neighborhood order because it requires more steps, but would consider joining a group order.
Making additional iterations after usability testing.
ITERATION 2: AUTO GROUPING
Simplifying the neighborhood order process & providing additional information
Auto grouping
After usability testing, it was clear this additional task of creating a group order was creating confusion and concern among testers, so from here, I decided to eliminate the option to create a group order by the user.
Instead, I moved forward with auto grouping, meaning users would opt into a neighborhood order but would automatically be grouped together after confirming their order based on three factors:
-
The proximity of restaurants they order from
-
Whether they order from the same restaurant
-
Proximity from each other to avoid extended delivery times
Neighborhood Order vs. Group Order Name
The name of the feature - “Neighborhood Order” - caused confusion among user so it was changed to “Group Order” instead.
ITERATION 2: VISIBILITY OF SYSTEM STATUS
Providing additional information to users at the discovery, check out and order updates stages of the customer journey.
Describe your image
Describe your image
Describe your image
Describe your image
EXPLORATIONS
Considering all users’ experience with Group Order
Before moving forward with final testing, I decided to question and explore a few additional options, including the ones below:
DASHER PROBLEM
Dasher is an independent employee who rarely receives a decent tip and needs to pay for their own gas.
They also deal with unexpected delays related to traffic, which results in fatigue. They would be less likely to consider taking a group order.
I found that Domino’s does a good job at optimizing delivery time by grouping people together and having one driver deliver the food.
EXPLORE USER FLOW CHANGES
UI EXPLORATIONS
Where is the best place for people to join a group order that would easily be a part of the flow and avoid confusion?
Explore different mobile patterns for tabs to avoid overcrowing on the restaurant page and consider visibility of system status, meaning keeping users informed on what this feature will look like. Provide more information than less to guide the user throughout the whole ordering process.
REVISED SOLUTION
Finalizing a solution: Order pool with DoorDash virtual kitchens
The final solution (for now) takes into account restaurants, the Dasher, and consumers for a solution with minimal impact on each of their experiences.
Grouping people automatically at the end, but only offering this feature for DoorDash’s virtual kitchens to avoid increased delivery times, continuous food quality issues, and Dasher pain points/concerns, while still solving for the pain point of high delivery fees.
Service fees would be reduced given the food would be prepped at a DoorDash kitchen.
Ideate
EXPLORATIONS
Considering all users’ experience with Group Order
Before moving forward with final testing, I decided to question and explore a few additional options, including the ones below:
Concerns with delivery time
Flow 1: User joins Order Pool through Restaurant page
Flow 1 caused more confusion than Flow 2 when it came to joining a group order because participants had to opt into a group order on the Restaurant page.
No incentive to create your own group order
Flow 2: User joins Order Pool through the Cart Page
Flow 2 did a good job at capturing the Order pool savings by giving users the option of go back and forth between screens.
ITERATIONS
Final changes from A/B testing (for now)
ITERATION 3: REMOVING ORDER POOL BUTTON
Revision 3.0
Discovered was that the reasons this flow was easier was because clicking on the ‘Order Pool’ filter button on the homepage made people believe that they had already entered the group order, so...
ITERATION 3: REMOVING ORDER POOL BUTTON
Revision 3.1
I deleted that button and added the category for order pools in progress and made it so people would automatically be added to an order pool if they clicked on the restaurant card from the order pools in progress category.
Clicking on the “Eligible for Order pool” category takes users to the list view.
ITERATION 3: ORDER DETAILS PAGE
Revision 3.2
The delivery information was added to the order details page due to questions participants repeatedly asked, “How is delivery time impacted by the additional people added to the Order pool after checkout?”
FINAL PROTOTYPE
Final prototype incorporates designs from flow 2 that were also beneficial to the user when placing their order and joining order pool. It targets users at different points of the flow.
Incorporate some of the components from flow 2 into flow 1. Flow one ended up being the best option where people immediately.
LEARNINGS
Three most important learnings from this project
Although I have reached a final solution (for now), this project showed me how complex it can get thinking of how one feature add-on can affect all other users, even in ways you’re unsure of, but that sometimes you can find inspiration for solutions from an unexpected place, such as Dominos in this case.
I’ve realized that sometimes inspiration comes from indirect competitors,
and companies in entirely different industries.
I see the value in a content design and design systems team to gather feedback about component designs and whether it fits into the current system, and how content writing is no small feat, especially when it comes to conveying important information in a succint and clear way where the user still feels supported throughout the process.
NEXT STEPS
Thinking about ways to continue to enhance the user experience
-
Conduct a final round of usability tests to see if these designs solved most issues uncovered in the A/B testing.
-
Partner with the marketing team to understand how to promote this new feature to consumers.
-
Try to figure out how to support food inventory from the virtual kitchens.
-
Work closely with the growth team to explore the possibility of a launch plan with dedicated dashers at these locations and establish a launch strategy such as launching in high populated areas.
-
Consider other alternatives such as adding this perk for DashPass users using Order Pool, or figure out if this would make sense there.