Redesign a database website for the ocean science community
Client
International Arctic Buoy Programme (IABP)
Research audience
IABP chief coordinator
IABP website data manager
Project length
10 weeks
(March 2023 - June 2023)
Team size
1 UX researcher (me) + 4 UX designers
Methods used
Contextual inquiries
Interviews
Competitive analysis
Card sorting
Usability testing
Developer review
IMPACT
-
Guided designers in making data-driven redesign decisions through 5 user interviews and contextual inquiries, 21 card sorting sessions, and 2 rounds of usability testing over 10 weeks.
-
The redesign and research significantly improved the website's functionality and user experience, as confirmed by the IABP data manager.
DOMAIN CONTEXT
If you've never heard of buoys, here's what you need to know before reviewing my case study...
Consider smartwatches: you may have used one or seen others using them to track health-related data for fitness, medicine, and more.
Similarly, buoys are the ocean's smartwatches. Researchers deploy them across global seas to collect critical data, such as air pressure and sea surface temperature. This data is invaluable for satellite validation, weather forecasting, and ocean science research.
​
Smartwatch
Buoy
BACKGROUND
IABP operates a network of over 200 buoys in the Arctic, providing daily data to the global ocean science community via their website.
The International Arctic Buoy Programme (IABP) is an international research organization that has provided meteorological and oceanographic data from a network of buoys in the Arctic Ocean since 1979. Currently, they maintain over 200 buoys in the Arctic. Their data has been used and cited in over 1,300 publications from 9 countries.
​
All IABP data are released and updated on a daily basis through their database website.
IABP website users
Researchers, Arctic enthusiasts, project managers.
PROBLEM
How can we redesign IABP website to enhance its data and information access for users?
-
Previous usability testing indicated that researchers had a hard time accessing and downloading data from the IABP website.
-
IABP also wanted to enhance their visibility to the public and potential investors to sustain and advance their work.​​​​
​
Challenge data view
Broken navigation
All daily buoy data from 2004 to the present were displayed on a single page, lacking search bars or filters.
​
​
​
Many important content pieces were hidden behind super links rather than a universal web navigation.
Time constraint
10 weeks
My team
1 UX researcher (me) + 4 UX designers
Outdated content
Much of the website content was outdated and inconsistent with IABP's goal of being a cutting-edge scientific data platform.
My role
UX research lead
CHALLENGES
A tight deadline and limited access to primary users were the biggest challenges I faced in deciding on my research approach.
Tight deadline
Minus the team formation and the final wrap-up, we only had 8 weeks to conduct all the research and finish design.
Limited primary user access
Arctic researchers are a less accessible user group, which indicated potential recruiting difficulty.
MY RESEARCH APPROACH
I aimed to ensure that designers made user data-driven redesign decisions, and enabled the new design to be evaluated by user data.
Prioritize! Prioritize! Prioritize!
With a tight project timeline, prioritization was top of mind for every team member.
As a UX researcher, I focused on ensuring that designers prioritized the right direction, guided by user data.
​
1. Established a participant pool and documentation templates
2. User interviews, contextual inquiries and competitive analysis
3. Card sorting
4. Usability testing
5. Developer review
1. Established a participant pool and documentation templates
WHY was this step necessary and important?​
Solved recruiting difficulty
Anticipating the challenge of recruiting primary users—researchers in a highly specialized scientific field—I proactively initiated participant recruitment at the beginning of the project. Leveraging stakeholder networks and snowball sampling, I ensured sufficient data resources for conducting primary research with this specific group.
​
Improved collaboration with non-researchers
Given the tight project timeline, I needed to collaborate with designers to conduct primary research. I used clear documentation templates to facilitate transparent communication about the research with non-researchers. These templates also helped our team track redesign decisions and monitor progress.
IMPACT
-
Allowed our team to efficiently conduct 5 user interviews, 21 card sorting sessions, and 9 usability tests within 10 weeks, collecting rich user data and guiding the redesign directions.
2. User interviews, contextual inquiries and competitive analysis
WHY was this step necessary and important?​
Complemented previous usability testing findings
Previous usability testing was conducted in a controlled environment and only involved researchers in the ocean science community. I realized our research needed to answer:
-
How do researchers use the website in a natural context?
-
Who else, besides researchers, uses this website, and how do they use it?
These questions are best explored through user interviews and contextual inquiries.
​
Gained design inspirations from best practices
Comparing the IABP website with similar scientific databases would help designers understand expected functionalities and inspire design ideas.
Since this method was closely tied to design, I had the designers lead this research piece and share insights during our user research finding share-out meeting.
IMPACT
-
Defined two distinct user groups and personas, covering both researchers and non-researchers in the ocean science community, based on their tasks when using the website. This became the focus of the redesign.​
​
1. Oceanographer
Regularly uses the website to search, view, and download data, with a high level of data expertise.
​
2. General Public
Regularly visits the website for IABP updates and contact information, with limited data expertise.
​
3. Card sorting
Why was this step necessary and important?​
Enhanced informational architecture to improve data and information access on the website
For both user groups, easy access to the data and information they need was a crucial requirement for the IABP website. The broken website navigation hindered this need, so our team decided to revamp the information architecture.
As a researcher, I ensured that the new information architecture aligned with users' mental models for accessing data and website information through 21 card sorting sessions.​​
​
Participants and sessions of card sorting
11 Ocean science researchers
10 unmoderated sessions on Optimal Workshop
1 moderated session
10 General public
10 moderated session
*I allowed participants to choose between unmoderated or moderated sessions to maximize data collection. Moderated sessions were more time consuming and involved researchers observing the card sorting process and asking follow-up questions afterward. Unmoderated sessions allowed participants to complete the activity on their own time via a provided link.
.png)
IMPACT
IMPROVEMENTS
For the general public:
Easily learn about IABP through the "About IABP" pages.
​
For oceanographers:
Easily access various regional data and formats through the "Arctic Buoy Data" and "Antarctic Buoy Data" pages.
​
For both:
Easily stay updated on IABP buoys and research through the "Buoy Specifications" and "Citations" sections.
-
Enabled designers to create a new informational architecture that simplified task completion for users, validated by later usability testing.​
​​
4. Usability testing
WHY was this step necessary and important?​
Evaluated design iterations with user insights
Usability testing is the ideal method to evaluate redesign directions and new prototypes, ensuring the design team addresses user pain points while meeting stakeholders' goals of attracting new users and retaining existing ones.​
​​​​
I GOT TWO IDEAS!
1. We had two weeks for usability testing and two user groups. Instead of testing both groups in one round, why not run two rounds of usability testing, one for each group? More iterations could mean greater product success.
2. The usability testing before the redesign project was conducted with ocean science researchers and included specific quantitative metrics. Why not reuse those metrics in the hi-fi usability testing besides qualitative questions to assess whether the redesign truly addressed user pain points?
​​​​​
​
Usability testing
PARTICIPANTS AND SESSIONS
QUAN METRICS
-
I adopted task difficulty levels and lostness analysis from previous usability testing when the user scenarios in our testing aligned with the previous ones.
-
I excluded the SUS scale used in previous usability testing because participants wouldn't have enough time during our testing to fully experience the website and provide a valid SUS evaluation.
IMPACT
-
Enabled the design team to iterate the prototype through two rounds of user feedback rapidly instead of just one.
-
Ensured the final design effectively addressed pain points for both researchers and general public.
5. Developer review
WHY was this step necessary and important?​
Increased the likelihood of the redesign being launched
IMPACT
-
Empowered the client to leverage our design and research to effectively address user pain points in a practical way.
RESEARCH FINDINGS AND IMPACT
Impact 1: Enabled data searches based on the most important buoy data types for users.
Previous usability testing and our user research identified that the absence of data search functionality was a significant pain point for researchers using the IABP website.
​
Data finding and discover available data both listed as serious usability issues in the usability testing findings.
​
- Finding from previous usability testing report.
"I actually only need one particular buoy's data, but there are no search option or other easy access for me to track that buoy. I have to go through all the table every time."
- Quote from user interview.
Two rounds of usability testing prioritized the data types important to both researchers and the general public, each with different search habits, enabling designers to create:
FEATURE 1
Buoy ID search bar on data table pages
FEATURE 2
Buoy data filters on data table pages
FEATURE 3
Simplified data tables with an option to "view more" if users need it
FEATURE 4
Buoy data filters and search bars on data map pages
IMPACT
-
Significantly reduced the difficulty for users in searching for specific buoys, as demonstrated in the final round of usability testing.
Impact 2: Provided grid and table data views to accommodate the different preferences of researchers and the general public.
User research revealed that the data view on the IABP website was challenging, even for researchers with strong data expertise.
​
"I have some research colleagues reporting to me that the data table was not so easy to view and use? It also took some time for me to teach them how to use it as well."
- Quote from user interview.
During ideation, designers developed a simplified grid data view to enhance the user experience.
Usability testing with the two user groups showed that the general public preferred the grid view, while researchers favored the table view. The team decided to keep both options: researchers would land on the data table by default, while the general public could easily switch to the grid view.
Switch view
Table View
Switch view
Grid View
IMPACT
-
Balanced the design to accommodate the preferences of both researchers and the general public, making the website more accessible to everyone in the ocean science community.
Impact 3: Simplified the data downloading process.
User research revealed that users struggled to download data because they could only download one buoy's data at a time, using specific Python scripts. This insight led designers to introduce a multiple data download function with a single "download" button and choices of diverse data formats.
I also consulted with the developer in the developer review to ensure that implementing this download option was feasible for the IABP website.
IMPACT
-
Resolved the user pain point of downloading multiple data sets simultaneously, significantly elevating user satisfaction as shown in the final round of usability testing.
"This is so exciting and much easier to use! Do you know when they plan to launch it, ready for us to use?"
- Quote from the final round of usability testing.
TRACKING RESEARCH IMPACT
My research and our redesign significantly improved the website's functionality, as confirmed by the client months after the project.
A few months after the redesign project ended, I revisited the IABP website and noticed that our design didn't seem to have been implemented. I emailed the client to understand why and to keep track of the project's impact. They responded with:
​​​​
“It's true that we haven't incorporated the design elements yet, but the research of you and your team did help tremendously with adding functionality to the site.”
—— IABP's website data manager
​​​
LAUNCHED FEATURE 1
"The data table now has two icons to allow the user to pop the data up in a new window or download the data."
LAUNCHED FEATURE 2
"The map is infinitely better now. It allows the user to go back in time, search for buoys, and zoom in."
REFLECTION
What I did well
-
Worked agilely and efficiently to complete the project within a tight timeframe.
-
Anticipated recruitment challenges and developed plans to address them.
-
Facilitated clear communication between users, designers, and the client.
-
Continually monitored project progress and the impact of research on designers and the client, refining my research approach accordingly.​​​​
What I would do differently
-
Established a UX metric to monitor website performance over the long term, providing a valuable research foundation for future iterations.
Other projects
Revamp the roadmap of clean energy mining trucks | Company: First Mode
User research, Research operations, Stakeholder communication
Find entry points for Gen AI to elevate employees' intranet experience | Client: Uber
User research, Design research, Usability testing