Covid Analyst
Product Design for an AI Powered
Regional Analytics for COVID-19
– Winner of three global hackathons
Role: Product Design, Interaction Design, Research, Visual Design | Timeframe: 36 hours | Tools: Sketch, Figma, Google Suite, Miro
Context
We are currently facing a coronavirus pandemic that is affecting everyone in unprecedented ways. With the desire to leverage technology and bring innovative solutions to challenges around COVID-19, our team – one data scientist, two engineers and myself – worked together in the Hack:Now 36-hour online hackathon. We aimed to build an impactful digital solution in a very tight schedule.
Problem Spaces
A day before the hackathon started and after some desk research, my team started to brainstorm ideas based in the challenges related to the current pandemic. We had two main interests: poor mental health caused by the “infodemic” as well as assisting vulnerable communities (e.g. Black Americans dying from COVID-19 at nearly 2.5 times the rate of white people). How could we use technology to solve for these issues? We started to ideate on possible solutions based on the questions:
How can we separate the helpful from the harmful when it comes to the pandemic news?
What does the situation look like in my own neighborhood? How about in my relatives and friends’ location?
Can we also provide a way for public authorities to reach populations who are disproportionately affected by COVID-19?
“Infodemic is an overabundance of information – some accurate and some not – that makes it hard for people to find trustworthy sources and reliable guidance when they need it.”
User demographics
I assumed that our audience would consist largely of adults 30-65 yrs old with access to the internet:
- Individuals and families looking to find credible info pertaining to their neighborhood
- Key persons working in national or local health technical institutes
- Community leaders and people from NGOs
- News professionals including online and social media outlets
Vision & Ideation
The main idea is to help people find trustworthy and relevant information about COVID-19 so that not only we are more empowered to weather the pandemic, but may also be able to help others in need. As a team, we envisioned a unified information hub where users could type a specific location and find most things needed:
Feature Prioritization
We had come up with various different concepts, but knowing that not everything could be done in such tight schedule and believing that starting out with less features was best, we determined to focus on three features that would add to a singular, solid product.
User Journey
Sketches
Two possible concepts are illustrated. The first consists of screens 1 + 2 + 3, and the second of screens 1 + 4:
1
The homepage should be very straightforward. It includes a search field and a help chatbot which will be accessible throughout the platform.
2
Second screen: after the user types their zip code, they have access to an interactive heat map and curated news relevant to their area.
3
Third screen showing the heat map with risk analysis. Below, graphs and stats help visualize more information besides what is shown in the map.
4
Next to the map, a card specifies the risk status and has links to more resources. Below, users can scroll though the curated news.
Mockups
Because the second concept was simpler and faster to build, I moved forward with it for the hi-fi screens. First, I did a quick visual analysis and checked out some reputable news sites for inspiration – John Hopkins University, Reuters, NPR, New York Times. I used toned-down oranges and blues to create a sense of optimistic seriousness, as well as a good amount of white space for a clean feel.
Usability Test
As we faced some time constraints, we chose to limit data to Wake County, North Carolina. I tested the app with four different users and got insightful feedback. Most users preferred that we included a brief explanation of what the product was in the homepage. Without it, it would be hard to tell what Covid-Analyst was just by looking at its name and the search field. The explanation also gave more credibility to the app.
Mobile Screens
My idea was to build a mobile-responsive platform to allow people to check updates on the go.
User Survey
After passing my mockups to the dev team, I started to work on a survey to be included in our app. This would allow us to gather more user data for the machine learning technology, and help us take a clearer picture of the effects of COVID-19 in different locations. In this survey, volunteers provide their zip code, ethnicity, gender, household size, and other relevant information.
Learnings & Outcomes
This project was challenging but an excellent learning opportunity. I got to collaborate closely (though remotely) with engineers and data scientists. Together we were able to ideate, scope and manage our work to build an innovative app in a very fast and effective way. Check out our product at DevPost.
Unexpectedly, Covid-Analyst won three prizes in different hackathons: 1st Place Overall Winners at the Resiliency Challenge, Track Winner: Health & Fitness at the World Hackathon Day, and Google Cloud Second Place at Hack:Now. We were happy to donate the money prize to the National Black Child Development Institute.
Our team worked in very different time zones!