Group 26.png

Yelp Redesign

Product Design, 2019



I heard a lot of complain about how Yelp is disappointed recently… so I decided to redesign the oldest flow of Yelp “Find a Restaurant” as my Verizon design exercise, trying to take a step back to evaluate the target users and capture some of the market they have lost in a new perspective.

Role Duration Tools

Solo Project 6 days Sketch, InVision

Group 49.jpg



Redesign the “Find a Restaurant” Flow of the Yelp Mobile to improve the experience based on a better understanding of the targeted user.


Provide two levels of solutions to the “Find a Restaurant” Flow. One in feasibility level to streamline the app and make it feel less overwhelming and more user friendly. The other in desirability level with a new feature added, to fill the missing gap inside this experience.


01. Early exploration


1. Background Research


Yelp mobile is an information heavy product. There are such abundant features happened in it with the pale experience which makes the user feel clutter, and the main purpose of the app is diluted.


To take a step back to evaluate the customers Yelp is trying to reach, how to best target and engage with them to create a user experience in the service design thinking that will differentiate them and allow them to capture some of the market share they have lost.


I decide to focus on redesigning the “FIND A RESTAURANT” flow of the Yelp mobile which is the original and main feature of the product, based on the latest data.




2. User research

There are New User vs. Returned User, Traveller vs. Resident in the Yelp. In the context of “Find a Restaurant”, I divided the users into five categories below.

Group 10.jpg


I prepared a list of questions from easy to hard and interviewed with eight people, and creating four personas below, to stress the goal and pain points of different types of the user.

Group 9.jpg


I asked the users to pin what they care when they try to find a restaurant into this Radar map, and summarized the results into two different user model, the “Quality Driven” and the “Location Driven

Group 32.jpg

3. Context Inquiry & Observation

The day I started this challenge happened to be the Valentine’s Day. So I asked the users to achieve two purposes, one is “Find a Restaurant for a Planned dinner”, the other is “Find a bar(or dessert) after the dinner without any plan”. I observed their actions and recorded the obstacles.

Group 7.jpg


-Questions 1:

Why prefer list view instead map view?


If I am searching in an area I am familiar with, and thus do not need the map (or I can look at the map location for a specific restaurant that interests me). I would rather just see the list since it shows a location’s distance (ex: .01 miles) from where I am.


-Questions 2:

Why google map in the end?


Once it knows my choice(destination), it can provide me the traffic information, add reminder for me and other sweet services.


4. Existed Experience Critique

Group 20.jpg

5. Competitive Analysis

From the user research, I got ideas about the competitive products the user using to find a restaurant. I conducted research on these products to find some insights.

Group 15.jpg

02. Research Analysis & Ideation

  1. Pain point & Insight

    I concluded into three main pain point from the research, the users currently deal with the Yelp in the “Find a Restaurant” Flow.

Group 34.jpg

2. Experience diagram

Based on research insights and pain points, I created this journey map to understand user better and create empathy with them, and land with opportunities.

Group 36.jpg

3. Opportunity Area

How might I???

Group 38.jpg

Frame the Design Problem


How Might I frame the “FIND A RESTAURANT” experience of Yelp mobile to better satisfied users with different goals?


03. Design Development

  1. Solutions

I provided two level of solutions to this challenge.

  • One is feasibility level, without new features added. I modified the following features based on the research, to streamline the app and make it feel less overwhelming and more user friendly.

Group 42.jpg
  • The other is in desirability level, with a new flow built in. I create a flow to let the users add restaurants into compare, and share with friends, to have the conversation over the decision together. Then let the Yelp know the user’s decision though “choose this” feature, to let Yelp prepare the “after decision” experience.

Group 47.jpg

2. Information Architecture


3. A/B Test

I conducted six small A/B test in the below functions before I landed into a entire flow

Group 59.jpg

4. Wireflow

Group 35.jpg

5. First-Hand Visual

Group 26.jpg

6. High-Fidelity Prototype & Iteration

I created a hi-fi prototype to test the two level solutions above and make modification based on the test result.

Group 29.png


I got other insights when I talked with the user after the testing, and I hope to develop and evaluate these directions in the future.

Group 74.png

7. User Flow

Group 24.png

Further Steps

  1. Modified the “compare flow” based on the insights from hi-fi test.

  2. Redesign the filter page into a more reasonable hierarchy with quick accessible filters

  3. Hi-Fi Test on the “Final Confirm” page to present expected outer connection for the users


Yelp is an information heavy product, and it already has trained a way that how user will explore the spots and write reviews. It is very challenging to redesign a very mature product like Yelp. My focus point is treating the whole service as a cycle, and treat Yelp as the key point inside this cycle, then fill the missing gap between the product and the user end. I am aiming at :

  • Modified the features based on user’s behaviors without changing the existed flow

  • Point out a new features based on the user’s potential need with a need flow add

Love this challenge, hope you too.