Quality data is key in the fight against climate change. The NSW Climate Data Portal makes enormous amounts of data available to the public — we're talking petabytes of climate data.
Once launched, the portal will be used by climate researchers, industry specialists, emergency services, government, the media, and the general public in order to better understand the state of climate change in the New South Wales, and thereby make informed plans for the future.
Leading the design effort
As the only designer on a team of about a dozen techies and government stakeholders, my responsibility was to lead the entire design effort on this project. I captured requirements from a team of stakeholders, analysed user research reports, and then delivered a prototype laying the ground work for the foundation of this big data portal. Following that, I user tested the proposed solution and delivered a report of findings and recommendations for next steps.
My work helped aligned the team of stakeholders on the vision for the portal, and it gave the development team guidance for the build.
No nonsense design philosophy
My goal was to keep it simple and stick as close as possible to the existing design system. This would ensure both a faster build, and a more accessible solution, since the design system components are accessible out of the box.
Built on open source
CKAN is the open source platform that's used for the technological foundation for this portal. After reviewing CKAN, I created a theme that utilised its strengths, and circumvents its UX shortcomings — of which there were a few. At the same time, the solution had to comply with the NSW Government design system, and of course not just tick these boxes but ultimately, and most importantly, meet the needs of future portal users.
As part of this work I spent some time mapping out a vision for a data visualisation feature. One that would allow people without data science degrees to use the site, and thereby make this important climate data accessible to a wide audience.
A group of 20 target users participated. The result was a 71% success rate on the tree test, and an 80% average success rate for key tasks in the click test — promising. While there's some room for improvement, overall I was happy with the results, especially on key data discovery tasks where the vast majority of users quickly found what they were looking for. In my report I recommended a more in-depth user testing effort once the site is in beta, but overall we're quite confident with our initial direction.
The turnaround time for this work was about a month — factoring in everything from project analysis, stakeholder meetings, a few iterations on the design direction, prototype creation and testing.