GoGriffith app

// UX design

The challenge was to identify a problem at Griffith Park caused by the COVID pandemic and design a solution. Through quantitative and qualitative research, we found that the uncertainty about crowd density was the most significant pain point for park visitors. To help solve this problem, we designed an app that showed visitors crowd density metrics that gave them the information they needed to make safe and confident decisions about visiting the park.

Screenshots of the GoGriffith app

Client. Academic project. SMC IxD Studio

Timeframe. 10 weeks

Responsibilities.

  • Stakeholder and user interviews
  • Visualization survey and prototypes
  • Storyboard and animatics
  • Wireframing and user testing prototype app

Research Methods

We used a survey of LA residents as a quantitative research method to get a general sense of people’s feelings about public spaces during the pandemic. Qualitative research methods, like stakeholder interviews, were conducted to gain a deeper understanding of more specific questions.

To identify the problems created by COVID-19 in Griffith Park we needed to talk to the stakeholders.

  • On site interviews of park visitors at the Griffith Observatory
  • Interviews with Griffith Park staff and park rangers
  • Interview with AP Diaz from LA Dept of Recreation and Parks
  • Interview with Dora Herrera from Friends of Griffith Park

To get a general sense of how the pandemic was affecting people who use the parks and other public spaces.

  • Created a survey for Los Angeles residents.
  • Distributed the survey thru local Reddit and Facebook groups
  • Conducted follow up interviews of survey respondents to get qualitative findings to augment the quantitative survey results.
Group picture at Griffith Park of people with masks onGriffith ObeservatoryCovid closure signage at Griffith ParkGriffith Observatory and the LA skyline at night
User interviews and observational research at Griffith Park

Research Insights

From our interviews, we discovered important insights that were reinforced with statistical information from our survey.

Due to the pandemic lockdowns getting outside is even more important, but people consistently find themselves in situations where it's too crowded and don't feel safe and comfortable when they arrive.

"We drove through traffic for an hour and spent 30 minutes finding parking. Once we got to the beach it was way too crowded. We just turned around and went home. The entire day was wasted."

Brock, Griffith park visitor
Pie chart 87 percent

Have you ever gone to a public space and felt uncomfortable or unsafe due to the COVID-19 pandemic?

Survey of 60 LA residents

People want information about crowd density because having a choice based on trustworthy information helps people feel safer and more confident in their decisions.

"I try to time when I go to make sure there's not a crowd. I always do an online search for any recent reviews that might give an indication but that's hard to find"

Michelle Jones, Facebook forum
Pie chart 89 percent

Would you be interested in knowing how crowded an area is prior to visiting?

Survey of 60 LA residents

We can utilize Griffith Park's existing infrastructure of security and wildlife cameras and combine it with currently available AI crowd counting technology to get real time measurements of crowd density.

Sunset at Mount HollywoodSecurity camera at the Griffith ObservatoryWildlife camera in Griffith ParkSurveillance camera footage with boxes counting peopleCamera footage showing proximity date and people count
Griffith Park wildlife cameras and AI crowd counting technology

Reframe the problem

How might we help people predetermine COVID-19 risk when visiting Griffith Park to minimize uncertainty and help them make informed decisions about the safety of their visit?

Persona

Once we framed the problem, we created a user persona that helped convey our research findings and design a solution.

Persona image of woman with two kids
Monica, 39

Since COVID Monica has had to juggle working from home with taking care of and homeschooling the kids.

Motivations

  • Get out of the house to de stress
  • Get the kids out of the house and away from the screens

Pain points

  • Outings with the kids take lot of time and planning
  • It's difficult to find out until they arrive whether a place is crowded

"We drove to through traffic for an hour and spent another 30 minutes finding parking. Once we got to the beach it was way too crowded. We just turned around and went home. The entire day was wasted."

Concept Storyboard

Our first prototype was a concept storyboard of how our persona could utilize our solution to help her uncertainty about crowd density at Griffith Park.

Storyboard sketch of a woman inside with her kids during covid
Monica is tired and stressed and needs to get out of the house.
Storyboard sketch showing crowd density information
She gets real time crowd information for different locations of the park.
Storyboard sketch of pinning an area
She chooses a location and pins it so it will be monitored for changes.
Storyboard sketch of 1 hour later
Storyboard sketch of mom leaving house with kids
On the way out, she gets a notification on her phone.
Storyboard sketch of phone with an alert
The location she pinned has gotten more crowded and she is able to change her plans and find a less crowded area.
Storyboard sketch enjoying a day at the park

Prototype testing. Visualization survey

There are many different methods to show crowd density data. We prototyped various data visualizations and created a survey to determine which ones were most useful to help users decide whether the park was too crowded.

What is the best way visualize crowd density data?

  • Which types of visualizations are most clear?
  • Which visualizations are most helpful for our users to make decisions about safely going to the park?
  • What made visualizations helpful or unhelpful?
  • The results informed which visualizations we chose and how they were used in the context of the app.

How we tested.

  • Research and prototyped different types of crowd visualizations
  • Distributed a survey via local Facebook and Reddit groups
  • Follow up interviews via Zoom for more in depth qualitative data
Screenshot of Google Crowd data visualization surveyScreenshot of Crowd Data Visualization survey

Visualization Survey Insights

We used the quantitative data from the survey and follow-up interviews to identify the correct mix of visualizations and help inform our design decisions.

Crowd density map

Plus icon
Sketch of crowd density map

"Shows the distribution of people very clearly and many people are already used to this design."

survey response
92%
of survey respondents
found this useful

How we implemented it.

  • easily scannable - used for primary visualization for location
  • overlaid with points data for added info and help color blind users

Crowd time graphs

Plus icon
Sketch of location density over time

"I look at the Google busy times graph. That's helpful to see how busy it will be when I get there."

survey response
84%
of survey respondents
found this useful

How we implemented it.

  • adds a layer of time to the crowd density information
  • added a graph to show crowd density for the week for long range planning
Screenshot of location screen from GoGriffith app

Images from security camera

Plus icon
Security camera footage with faces covered

"I am used to looking at a camera feed and it is reassuring to know it's real time. You can see how people are spaced out"

survey response
93%
of survey respondents
found this useful

How we implemented it.

  • used to supplement crowd map
  • reinforces trust in the date
  • covered faces to address privacy concerns

AR/VR crowd simulations

Plus icon
Sketch of augmented reality crowd density

"The 3D and AR simulations are better for when you're at home looking up data. The 2D maps are better for looking on the fly."

survey response
67%
of survey respondents
found this useful

How we implemented it.

  • users found the feature interesting but one they wouldn't use much

Prototype testing. Go Griffith app

We created a low-fidelity digital prototype to test both the flow of the app and the legibility of the crowd density information.

What are we testing?

  • Does the user understand the purpose of the app?
  • Can the user navigate the primary user flow?
  • Does the app provide enough crowd density information to allow the user to make decisions about visiting Griffith Park?

How we tested.

  • Ideate with sketches
  • Clickable prototype with Adobe XD
  • Remote user testing with Zoom
  • Performed 2 rounds of iteration and testing

What did we learn?

  • Needed clearer onboarding to introduce app and functions
  • Crowd density visualizations needed more information and keys to explain what they represented
  • Confusion about process of pinning and pin settings
  • Users wanted to plan trips and see crowd information for later in the week
  • Overview map icons were unclear and colors need explanation
Sketch of app wireframes
Ideation sketches for pin page
Screenshot of remote user testing with Zoom
Remote user testing via Zoom
Screenshot of first prototype overview map
Overview map
Screenshot of first prototype location info page
Location info page
Screenshot of first prototype of pin setting
Pin Settings
Screenshot of first prototype pin set and share screen
Pin set & share

Final Prototype

Wireframe of GoGriffith final prototype

What I learned.

What is the least amount you need to properly tell the story?

  • With the amount of information and insights we had to share, it was difficult but crucial to boil it all down to the bare minimum for the audience to understand the main ideas of the project.
  • Brevity in storytelling is a weakness of mine, but one that I am aware of and am consciously trying to improve on.

Properly mixing the use of quantitative and qualitative data provided the most valuable insights.

  • We switched between quantitative (surveys) and qualitative (interviews, user tests) methods depending on the situation and allowed the findings from one to influence the other.
  • Results from the initial stakeholder interviews informed what type of questions to ask in the surveys, and results from the survey gave us data to dig deeper into in our follow up interviews
  • In general, the surveys gave us concrete answers to simple questions (example. 92% of people found crowd density maps useful) while follow up interviews helped us understand why (the crowd density maps were easily scannable and the format was familiar).

Prototyping and testing is not just for wireframes.

  • By creating a survey to test different visualization prototypes we were better able to understand how people digest crowd information to make decisions.
  • In follow up interviews, we were able to understand why certain visualization were useful. This gave us useful information for designing and iterating different types of visualizations and how to present them in the app.

The COVID pandemic forced us to adapt and create new UX methods.

  • The lockdowns created new constraints such limiting our ability to meet and work with our teams and stakeholders in person and do on- site observational research.
  • Through adaptation and trial and error, we learned how to use new remote work arounds and strategies that will be useful tools in the post-COVID world.