The seattle.gov homepage is visited by 77,300 residents every week on an average. According to usability studies conducted by the city in the last year, residents have reported experiencing fatigue and confusion looking for information on the city website.

• The website content is currently organized based on the internal departments existing in the city government.
• The residents are unaware of the departmental structure within the city government and so the navigation experience is not intuitive for them.
• Residents miss out on beneficial resources offered by the city because relevant information is not easily discoverable.

To rethink the taxonomy it was important to understand the resident’s mindset when they land on the website. Based on past research we know they came with a goal in mind and this goal differed based on the type of resident. To make sure we capture all needs we took a personified approach by consulting the personas previously created by the team for city residents.

Next, we tested if our sitemap accommodated the mental models that the residents have when they come to the city's website with a goal. We used Userlytics to perform unmoderated usability testing with participants from each of our persona segments.

• Card sorting x 10 participants
• Tree testing x 9 participants
• Testing with low-fidelity prototype x 6 participants

Before each exercise, I suggested we ask the participants what each category means to them and what they think they might find in that category. This helped us understand their assumptions about the terminology detached from any bias that our card sorting/prototype might put forward.

1. Government Help was the default choice for participants who did not know where to place a card. To be more indicative of its actual content of discount programs, it was changed to Government Aid.
2. We realized that Data and Research did not have much chance of being used by the average resident and eliminated it as a high-level category.
3. Learning & Education and Transportation & Development were two new categories I pushed to add as they were prime concerns for multiple participants when asked what they would expect to find on a city website.

While categorizing, the “by audience” approach seemed more in alignment with the city personas. However, we often ran into a situation where a resident could identify with multiple audiences. Therefore, we decided to go further with the “by topic” navigation model while the “Information by audience” as a subcategory instead.

Once we had a good hold on the taxonomy we started brainstorming ideas for appropriate navigation elements to improve the UI. We went through the first round of sketches to showcase our ideas and vote as a team. We concentrated on three key design areas in our sketches:

1. DESIGNING A MEGA MENU.
   The taxonomy that we had come up with had 8 top-level categories that had multiple sub-sections. To accommodate this structure, each of us sketched iterations for a mega menu.
2. STRATEGIC PLACEMENT OF TOP TASKS.
   Since the taxonomy was fairly comprehensive, the team thought it would be a good idea to highlight the top tasks of each category.
3. AGGREGATE PAGES TO CONSOLIDATE SIMILAR INFORMATION.
   These pages would serve as the hub for a set of similar services that span across departments and are united by a goal or a particular audience.

Next, we recruited residents that qualified as any of the city personas to try out the new navigation. We ensured that each persona was represented in the participants. We conducted 12 moderated in-person usability testing sessions.

The results of the usability testing showed the improved time on task metrics and a much more intuitive experience looking for information. After discussions with the departments and sharing the results the roadmap for the first implementation of the new redesign has been set.

• Through this project, I had an opportunity to work on information architecture on a large scale.
• I learned to design for equal representation of multiple target users i.e Seattle residents with varying needs.
• As a next step, I would track user interaction with the website and do further usability testing to iterate on the design.

A second goal for this project was to ensure all designed layouts are accessible as required for government websites. To that end, I created a set of accessibilty guidelines in the form of a checklist that can be used by designers & developers to ensure they are meeting the standards. This checklist is based on the WCAG 2.1 guidelines and is AA compliant.