Resy Personalization Engine
Proposing a personalized recommendation engine to design an enhanced restaurant reservation experience.
Summary
Role
As new interns at
Konrad Group, our cohort was split into teams to complete a 2-week-long project. As part of the onboarding process, the "mock project" allowed us to practice using design thinking from start to finish, before being put full-time on client projects.
As the sole designer on the team, I was responsible for actualizing and prototyping our final concept. I worked closely with the consultants as we constantly iterated and came to research-based solutions.
The Team
This project was a team effort, and each member contributed their expertise to produce a final product that we were all proud of. Hailing from different schools, degrees, and backgrounds, each of our unique perspectives was integral to the development of the project.
The Context
Resy is an online platform where users can discover and book restaurant reservations. The website and mobile app are widely used across America, especially in New York City, where Konrad Group is based.
Given the brand's rapid growth in the past few years, now is a pivotal point for Resy to prove that they can keep up with user's high expectations.
Our task for the two-week-long sprint was simple:
Improve the web and app user experience for two key personas in order to unlock additional platform growth in NYC
Research
Resy's Current State
Resy's current state leaves significant room for growth and opportunity.
Fragmented User Experience
The web and app experience are vastly different - the website features tailored picks and clearly organized lists, whereas the app is impersonal and limited to ambiguous results.
Low Brand Loyalty
Nearly all interviewees indicated a lack of strong preference for using Resy during the search and reservation process, instead using it, and other competitors, as a means to an end to make a reservation.
Uncompetitive Market Share
Resy only takes up a small portion of the restaurant reservation space - only 10% among a sample of restaurants in the US that take reservations compared to OpenTable’s 60%.
Key Pain Points
After just a quick scroll through the current Resy experience, we found ourselves frustrated and confused. We identified a few key pain points that we wanted to address in our solution design.
- Lack of tailored results
- Low information transparency
- Inaccessible navigation
- Cluttered profile page
Key Pain Points
After just a quick scroll through the current Resy experience, we found ourselves frustrated and confused. We identified a few key pain points that we wanted to address in our solution design.
- Lack of tailored results
- Low information transparency
- Inaccessible navigation
- Cluttered profile page
User Interviews
We conducted eight user interviews across a range of demographics, eating habits, typical spending, and familiarity with Resy. Through this process, we empathized with our users and learned about their pain points.
“I always struggle with finding a restaurant … I like how other platforms use order history and favorites to recommend restaurants”
Tech-Confident Consultant
“Resy lacks personalization features … it should be able to quickly determine where I want to go based on previous bookings”
Resy Power-User
“It would be nice if I could put my allergies, which are more than just preferences, and have them expedited to restaurants”
Allergy-Wary Designer
User Personas
We landed on two user personas that represented a large portion of Resy's client base, and focused on creating solutions with Busy Bella & Refined Romulus in mind.
User Journey
Through considering a user journey for both of our personas, Refined Romulus and Busy Bella, we tracked low and high points of the user experience.
High Points
- People are thrilled when they find their restaurant and make a reservation
- People often indicate special requests and accommodations on the platform
- People appreciate a seamless dining experience at restaurants
Low Points
- People don’t enjoy finding restaurants using current discovery and search features
- Choosing between multiple restaurants without curated results is frustrating
- People want more transparency around notifications
Defining the Problem
After gaining an understanding of who our users were, and more importantly, the experiences they wanted from Resy, we started to hone in on one central problem.
Resy users felt overwhelmed by the amount of restaurant options they were looking at, which led to frustration and inability to make a decision.
The Paradox of Choice
When confronted with an abundance of options, people experience stress over making the right choice.
In a famous study from 2000, researchers set up tables at grocery stores, offering samplings of various jams. On one day, the table had 24 jars of jam, and the next day they only offered 6 flavors.
They found that buyers were actually more likely to purchase, and feel satisfied with their choice, from the table with only 6 options.
Our Resy users don't want to look at hundreds of unorganized restaurants, because choices are easier to navigate when there are less options. They would be more likely to decide on restaurant if they were presented with more personalized options - they would be more likely to find their perfect jar of jam.
Identifying a Solution
So, how do we help our Resy users navigate the stress of making a reservation, without overwhelming them with too many options?
By personalizing the reservation process, Resy can minimize the cognitive load for users, making the reservation process stress-free and efficient.
Enhanced Onboarding Flow
Our new onboarding flow gathers data on three key topics:
- Preferred Dining Categories
- Dietary Restrictions
- Special Accomodations
Personalized Explore Page
- Houses all curated Recommendations
- Offers filtering abilities
- Enables users to discover new favorites.
New User Profile
Our new profile page lets users find their reservations, hit list, notifications, dining preferences, and settings all in one place.
- Users can view past and future reservations
- View and edit Hit List of want-to-try spots
- Notifications, settings, and dining preferences
Emphasis on Transparency
- Explains how user data is being used
- Legitamizes reservation reccommendations
- Promotes feeling of trust between users and Resy
Arjun
2nd Year Symbolic Systems Student, Stanford University
- Feels overwhelmed by LinkedIn
- Unsure of his career path; wants to meet people with similar interests to discover different post-grad careers
- Made a group chat for students in his major, but they didn't have everyone's phone numbers
- Too many channels in which his professors and classmates communicate, formally & informally
Serene
3rd Year Human Bio. & Society Student, UCLA
- Uses LinkedIn to reach out to alumni from her school; some alumni don’t want to be contacted
- Wants to meet alumni in her field who are excited to mentor her
- Too many connection requests from classmates she doesn’t know and that aren’t in her field of study
- How can she connect with more relevant students from UCLA?
Key Performance Indicators
The success of our personalization feature set will be measured by five benchmarks.
Full Site Navigation
Home Page
Here's why this page works:
- Nod to comfortable design systems
- Clear tags to classify alumni
- Focus on important events
Account Page
Here's why this page works:
- Saved content easily accessible
- Page for personal organization
- Quick links to other pages
Final Thoughts
This was my first time working working directly on a project with a team of strategy consultants. It was incredibly useful to understand the way they analyzed the problem from a business and data perspective. Each of us offered different approaches to how we would begin solving for a solution.
As the only designer on the project, I worked in parallel to the consultants as we identified the problem and solution space, and made decisions about how the interface should look and feel. When it came to prototyping, I actualized our wireframes into high-fidelity designs, and made adjustments to our original solution requirements accordingly.
I truly believe the best teams are ones where no two members are the same, and that was definitely the case for ours. It is experiences like these where I learn the most, and I am extremely proud of the product we produced in a short time frame.
Tap through our prototype to see how it works!
Embedded Prototype
