Enjoy every outdoor experience with your dog

HappyPath introduces dog walking mode in way finding and social interactions. Urban dog owners often encounter unexpected interactions when they bring their dogs to public areas. Finding a non-disruptive route and managing their dogs' behavior in front of others are their big pain points. HappyPath helps users make informed decisions in route planning and effectively communicate about their dogs' temperament with others. Users can now avoid unwanted interactions and have a happy journey with their dogs!

Find non-disruptive routes to walk your dog

HappyPath helps you find unexplored dog-walking spaces in the city, or make adjustments to your frequenlty visited routes to avoid unwanted disruptions.

Inform others about your dog's temperament and social needs

Let's inform other people of your dog's willingness of social interaction by editing and sharing your dog's bio. You can set your dog's social status before or on the go.

Get alerts about cautious areas along the way

Worried about running into a crowd and making your little animal too excited or frighted? HappyPath keeps pushing to you crowd-sourced information about cautious areas when you approach. Stay informed, stay prepared.

My Role
UX Researcher

Designed a semi-structured interview, and conducted 6 interviews with urban dog owners.
Designed part of survey questions.
Analyzed user research data and synthesized findings in personas and a "before" journey story.
Designed the user tasks of heuristic evaluation.
Designed usability benchmarking tasks and conducted 2 in-contest testings.

UX Designer

Created paper mockups, low-fidelity designs to communicate about design ideas.
Created the information architecture of the entire product.
Iterated wireframes addressing problems that occurred in the heuristic evaluation.
Designed high-fidelity UIs following the iterative process.
Created an interactive prototype for user testings.
Redesigned some features after evaluation to improve product usability.

Understand the Problem

According to the 2015 - 2018 APPA National Pet Owners Survey, 68% of U.S households own a pet, which equates to 86.6 million homes1). Our team members are all pet-lovers. We have all heard about the inconvenience of living with pets in the city. We chose to design for this special group - "urban dog owners" in the hope of making their urban lives more enjoyable.

Secondary Research

To have a broader understanding of the background, we did secondary research covering pet industry, pet health, and urban lives. After reading relevant articles, reports and papers, we found an opportunity space:

Due in part to the close quarters of urban life, almost all urban dogs are “indoor” dogs. More than 70 percent2) of dog or cat owners living in urban areas agree their pet sometimes has anxiety/stress issues. Only 55 percent of dog owners living in suburban or rural areas say their pet has these problems.Most of them recognize that in many ways cities are not ideal environments for dogs, and that they have to take additional steps to manage their dog’s stress and safety.

What does your dog's day look like

One of the major struggles of urban dog owners is to give their dogs enough 'outdoor time' while managing full-time work and study and the long commute. Otherwise, the locked-up homes and lack of green spaces build up the animals' anxiety.

User Interviews

To have a better understanding of where our user group stands, we conducted 12 semi-structured interviews in Atlanta City. We chose to do semi-structured interviews because it establishes consistency across all data collected, but still allows us to prioritize some questions and drill down when interesting topics come up.

I designed the interview questions with a teammate and drafted an interview plan [View Document]. The team visited two places where dog owners frequent. Each group had one person acting as interviewer and the other as a note taker. We switched our roles in the middle of it.

Below is an outline of our interview broken down to 4 main parts.

What we heard...


We uncovered 3 major themes that were consistent across all 12 participants.

Walking/exercising dogs in public without encountering negative interactions.

Make sure dogs receive consistent treatment from different caretakers or strangers.

Work-Dog Balance


At this point, we still had a complex problem space. It was not clear to us which theme had the biggest potential, and it would be too ambitious to try to address all three. So we decided to collect some quantitative data through online surveys to point out a direction for us.

We used convenience sampling to collect the data - surveys went out through social media posts, direct messages and word of mouth. Atlanta is a dog-friendly city so lots of our friends and families had dogs at home. We were able to collect 32 responses after discarding 1 invalid response.

Research Data Analysis

The goal of this survey was to collect demographic data to help us decide which problems were proritized.

We analyzed research data in two sessions: an 1-hour affinity mapping workshop (we combined the interview results) and another 1-hour white board discussion.

We summarized the problems and whys, user's goals and needs. Finally we were able to conclude the problem to one statement:

The Key Problem

Dog owners find it difficult to take their dogs out while keeping dog's behavior and emotion stable, due to interference from other people and dogs that are not aware of the current training, nor of the dog's temperament.


After we defined our problem, we synthesized data and findings from the previous "empathy" research by creating the following personas. These personas put together a miniature of the urban dog owner group and the problematic environment. Now the team became more engaged with our user group by picturing them "real".

Roll Back The Story

We drew a storyboard to reflect the conflict between the 4 roles in our persona. By rolling back the scenario where the defined problem took place, it helped us root design thinking in the users' stories and produce involvement and insight.

Design Objectives

By engaging ourselves in the users' story, we came up with a clear design objective: Our goal is to help urban dog owners enjoy taking their dogs out by avoiding unwanted interactions from other people and other dogs. We want to orbit the ideation around this design objective, which involves:

Help dog owners choose destinations to minimize negative interference
Communicate with others (other dog owners, non-dog owners) about the dog's temperament and how to interact with their dogs.


The team had a brainstorm to kickstart the ideas. We chose to leverage the two phases of group problem-solving, divergence and convergence. By diverging, we individually generated as many ideas as possible before converging and using affinity mapping to discuss and narrow down ideas. Through creative thinking and brainstorming, we were able to get creative with design ideas and to maximize the unique contributions and ideas of the entire team.

Brainstorm Analysis

Some ideas were very similar in their approach to the problem. I abstracted three themes and grouped our ideas in their corresponding domain: Status Awareness Device, Online Resources, Training Progress Management.

We then analyzed each idea by thinking about the resources it takes to implement and talking about what we liked and disliked about it. We took out some ideas and sorted the better ones by feasibility and clusters of affinity.

How Did We Land On HappyPath?

After the initial ideation phase. 3 ideas were developed further and presented to people outside of our team.

We consulted some dog owners and experts who had significant experience designing dog products to give us some expert insights.

Considering the difficulty of putting technology on dogs and the applicability for different dogs, we abandoned all wearable ideas. Some dog owners and 1 expert expressed their interest in exploring more dog-friendly paths in the city. We recognized the potential in "HappyPath" and decided to pursue this direction. We found that some other good ideas emerged in our brainstorm can also be integrated into HappyPath to make it more powerful.

What is HappyPath?

HappyPath is a mobile application that allows users to explore parks, trials, other dog-friendly places in the city and provides navigation and information services. We incorporated crowdsourcing features to deliver to users the most updated information about the destination and its transportation condition. Status awareness setting allows users to communicate about their dog's temperament, behavior pattern and status to the public.


Information Architecture

Based on the paper sketches that all team members worked collaboratively on, I made an information architecture to keep all features in track. (blue boxes highlight corresponding user interactions on the page).


It's time to make some mockups! With the product structure and corresponding user tasks fleshed out, I designed the following UIs in low fidelity.

How did we get feedback?

We wanted to get user input before diving deeper in the design. Cognitive Walkthrough is a very efficient way to reduce the number and severity of design errors that will be discovered in the later user testing. It's more valuable to have the evaluation result before our design is complete.

User Flow

I created this user flow specifically for cognitive walkthrough. The tasks focus on important user tasks in a typical use case. Secondary features such as checking fun activities in the city and quick start navigation to favorite places were left out in the main flow.

Getting Feedback

We conducted 7 cognitive walkthroughs with dog owners. Each test lasted about 20 mins. While they went through the typical user tasks, we asked them to think aloud and recorded their feedback. I then marked the issues we thought were important and needed to be redesigned on the wireframe.

Issues Revealed
Wireframe Modification

After that, the team discussed potential improvements. I redesigned the following pages to address the problems.

High-fidelity UI Design

Finally, we solved all the issues. I then designed a high-fidelity prototype. Below are several demos of the main use cases (please check if your browser supports video tags).

Tools: Sketch, Principle

Explore Potential Destinations
Make Informed Decisions
Interact with Other Dog Owners
Ways to Change Your Status
Identifying A Report
Share Resource and Fun Things

Integrating with existing ecosystems

Because we received a lot of comments where people stated they'd like to use "HappyPath" features on the navigation app they were already using. It led us to design the Google Map plugin version of "HappyPath".

Additionally, we acknowledged that there are actually many benefits of incorporating HappyPath feature to Google Map:

Google Map has a large user base. Because user-generated information is one of our key resources, it would be hard to get the first group of users otherwise.
Google Map can help filter the accurate crowdsourcing information, with more users rating the report and falling under the platform's supervision.
There are opportunities to extend our partnership with local business (such as dog cafes, dog-friendly restaurants, botanic gardens) that are already partners of Google Map.

..So, what will the Google Map plugin look like?

Considering probable limitations, the Google Map Plugin was a clipped version of the standalone app. We took off the "Fun Events" and any point mechanism to keep it simple. We also took away the "wave" feature because we considered it optimal to limit user interactions in such a populated and complex environment.


It's great we have two versions of HappyPath: a standalone app and a Google Map Plugin. Now it's time to evaluate how well we addressed the problem. We did 4 preference tests (in the lab) and 5 in-context usability benchmarking in total with dog owners in Atlanta.

Prior to usability benchmarking, we conducted preference tests to figure out which version was preferred and thus can be carried into the usability benchmarking. The preference test also served as a pilot study of our evaluation plan. By analyzing the test results we were able to modify several flaws in our benchmarking tasks.

The preference test result indicated more users prefer the Google Map Plugin version. So we conducted usability benchmarking with the plugin version.

Our usability benchmarking received positive feedback. The success rate for all tasks were 100%, which partially benefited from the tooltips we added on the launch screen. The overall SUS score was 77.5, which we interpreted as above average usability.

Use Scenario

Post-evaluation Modification

In the evaluation tests, we observed several problems that were consistent among different test subjects. I made a few modifications to address those problems.


Looking back at the process, we’re proud of our cumulative efforts on the project. I also learned a lot of things:

There will always be trade-offs

We debated a lot on whether allowing users to report custom message, and if not, what are the types of report that make sense to users. Custome message, on the pros side, will certainly provide more dynamic information to users, but on the cons side, it risks environment cleanness and raises a lot of concerns of language. In this battle between flexibility and security, we chose security. I admit we could have down better if we did another survey focusing on gathering the types of useful information, but given the limited time and resource, it was hard to come up with the right set of option.

Social features are tricky

We recognized there was an opportunity of dog owners social interaction in the ideation phase, and tried to touch on that by introducing the "wave" feature. However, we received mixed messages about how users think of them. In our preference test where the app version had "wave" feature and Google Map plugin did not, some users liked the cute and simple interaction, but some others said they wanted to mute this because they didn't want to receive "wave" from others.

Listen to our uses

We spent a lot of time designing pleasing UIs and smooth interactions for the standalone app, but the simplified Google Map plugin won user's favor in preference tests. We were glad that we listened to our users and made that second version. It would otherwise be a pity that you put a lot of thought into a product and no users download it.

Question: How to deal with a "half-and-half" dilemma?
Take the custom message as an example, half of the users liked it and the other half did not. In this situation, if you win 50%, you lose the other 50%.

Question: Where is the boundary of mobile devices?
In this project, we developed a digital solution that resides on users' phones. Now people will listen to real-time reports while going out with their dogs, is it a good or bad thing? Should we encourage people to go "phone-free" in such activities?

We could have done better!

We let us go down the road of running users researches and making evidence-based decisions. Every piece of decision was backed-up, but when I look back, there were some flaws: The app had a retention problem in its nature, where it provides less value to users once they have explored all the dog-friendly places in the city. We worked hard to solve this problem by introducing the "crowdsourcing" feature which constantly generates up-to-date information. But I think we could have done better if we took a more holistic view earlier in the product design phase.