[Hop-In]
Making spontaneous dining effortless by connecting cravings to open tables through live wait times and AI-guided recommendations.
CONTEXT
Scenario
It’s Friday night and you’re craving something now! Sushi, pizza, anything good. But your usual apps only show late reservations or vague “walk-ins welcome” labels with zero insight into actual wait times. You’re stuck guessing, risking a 60-minute wait, or wandering from place to place hoping for a table. Before long, your exciting Friday night plans shift from anticipation to frustration, all because finding a place to simply hop-in is harder than it should be.
The Challenge
Today’s dining apps work great if you book days in advance, but fall apart the moment you want food now. With little real-time insight into walk-in wait times, spontaneous diners are left guessing availability, gambling on lines, and wasting time. This revealed an opportunity to bridge that gap by enhancing real-time walk-in visibility and introducing an AI host that guides users from cravings to nearby options they can actually get into.
SOLUTION
Live walk-in availability and AI-powered dining that connects cravings to open tables
By combining real-time walk-in wait times with an intelligent AI host, Hop-in helps users cut through the guesswork. They can instantly see which restaurants have space and receive personalized recommendations, with the option to join a waitlist or notify the restaurant in one tap.
Explore/Search Results Page & Add to Waitlist
The updated Explore page shows nearby restaurants with live walk-in wait times. Pick a spot, join the waitlist instantly, and get real-time updates as your turn approaches.
.png)
AI Host
The AI Host uses dynamic, conversational cards that are personalized based on user prompts, surfacing the most relevant restaurant details on dedicated cards for easy browsing.
.png)
SOLUTION
RESEARCH
“Walk-Ins Welcome?” Lack of walk-in visibility and reservation prioritization create barriers for spontaneous diners.
I began my research through a competitive analysis of the current top apps in the restaurant reservation space. With the use of AI growing I decided to utilize its capabilities to help outline the key competitors and understand their feature capabilities to then identify respective gaps highlighting key feature opportunities for [Hop-In].
AI Prompt
You are a UX Researcher tasked with analyzing competitors in the online restaurant reservation space.
-
Identify 2–3 major competitors.
-
For each competitor, summarize their app’s key features.
-
Evaluate their strengths — what they do well from a UX and usability standpoint (e.g., interface design, personalization, efficiency).
-
Highlight their weaknesses or gaps, especially areas where the user experience could be improved.
-
End with a comparison matrix
Once I understood the key competitor gaps and brainstormed a couple opportunity areas, I decided to do a deeper dive into user sentiment. Thus, I began by looking through a variety of dinning articles to see how individuals felt on the current dining environment. After a couple articles I was noticing a clear pattern emerge: spontaneous dining has become increasingly difficult. Users shared that popular restaurants were often fully booked days or even weeks in advance, forcing them to plan meals far ahead of time. Those who attempted to walk in described long, unpredictable wait times and inconsistent information about availability.
The more people I spoke with … the more I realised that the spontaneous fun of walking into a restaurant had given way to the frustration of realising the need to have made plans weeks — if not months — ahead of time.
My boyfriend asked how long the walk-in wait would be, and the manager told him anywhere from a minute to 2.5 hours. … That had to be some of the worst customer service I’ve ever witnessed.
Peak hours at popular walk-in restaurants can lead to extended waiting periods, which can negatively affect customer satisfaction … The absence of reservations means that customers may need to wait in line for hours, which can be frustrating and inconvenient.
SOLUTION
My next step was to create a user journey map illustrating the current walk-in dining experience, pinpointing the key pain points and finalizing feature opportunities that would guide the design of [Hop-In].
Culminating my research, I identified two major gaps for spontaneous diners: they can’t see real-time wait times, and they have limited ways to surface options that match their preferences in the moment. These insights shaped [Hop-In]’s direction, leading to features that pair live walk-in availability with an AI host that helps users find open spots aligned with what they’re craving.
DESIGN PROCESS
Designing with spontaneity and personalization in mind.
With the core pain points defined, I moved into the design phase by creating low-fidelity wireframes to map out the apps flow.
.png)
I decided to focus on the two main features that granted the greatest design impact based on user painpoints: The Home/Explore page and the AI Host feature​​
Explore Page & Search Results
The Explore page serves as the main search screen giving users immediate access to nearby options along with deeper, at-a-glance details like ratings, cuisine, distance, and real-time walk-in wait times. These designs focus on clarity, usability, and effortless comparison to support confident, spontaneous dining decisions.
_edited.jpg)
Explore-Page
_edited.jpg)
Search Results Page
AI-Host
While testing existing AI tools with restaurant-related prompts, I found that most delivered long, repetitive responses that only slightly changed when prompts changed. Co-pilot offered a more visual approach with interactive cards, but the content itself didn’t adapt meaningfully to different user asks. This highlighted a clear opportunity: users don’t need static paragraphs, they need fast, context-aware options that shift based on their intent. This insight directly informed the design of the AI Host, which uses dynamic, conversational cards that update to reflect the user’s specific cravings, context, and availability.
AI Prompt 1
Find me a cozy place with great cocktails and under a 20-minute wait for 3 people.
.png)
AI Prompt 2
Show me a couple highly rated cocktail lounge style places walking distance from me.
Both ChatGPT and Copilot produced dense, paragraph-style responses that made it difficult to quickly scan multiple restaurant options. While the content slightly changed based on the prompts, the shifts were subtle and hard to spot, making it challenging to follow the thread of the conversation or feel any real sense of personalization. This gap informed my approach: instead of long text blocks, I designed dynamic, conversational cards that highlight the key details the user actually asked for. These cards update by analyzing user prompts, surfacing the nuances of each prompt in a scannable, interactive format. Streamlining the back-and-forth and helping users reach a clear decision faster.
Other Card Examples Based on Prompt Inputs
_edited.jpg)
FINAL DESIGN
REFLECTION
O1
O2
The Challenge of Making AI Feel Useful, Not Overwhelming
Working through this project helped me realize why I never used AI to choose restaurants: most tools respond with long, generic text that’s hard to scan and rarely adapts to what you actually asked. This insight became a core design challenge for Hop-in: how can AI feel genuinely helpful rather than heavy? My approach focused on shifting from paragraphs to small, digestible pieces of information, surfacing meaningful insights through dynamic, conversational cards that adapt to what the user asks.
​​​​​
Designing for Spontaneous Users vs. Planners
A key learning in this project was understanding the very different needs of spontaneous diners compared to planners. Planners typically browse menus, read reviews, and make reservations weeks in advance, so they’re more comfortable with detailed information and longer decision cycles. Spontaneous users, on the other hand, want fast, scannable details they can act on immediately. By focusing on clarity, brevity, and at-a-glance usability, I was able to tailor the experience to support quick decision-making without overwhelming the user.