The mission of the Queens Economic Development Corporation (QEDC), a non-profit, is to provide local entrepreneurs with resources and guidance to start, promote, and maintain successful small businesses. After yearlong restrictions directly impacted the community and its local businesses, QEDC’s mission matters more now than ever.

Its website, however—the main way to connect with QEDC and access its resources—was failing the organization's mission and the people it serves, and the team at QEDC knew it.

“The website is large and chaotic because we keep adding more pages," Michael Gillen, QEDC’s Director of Technology and Communications, told me. "So it's confusing and not as easily accessible and understandable as we would like. We want people to go on the site and immediately be able to learn how we can help, access our programs and resources, make an appointment with an advisor, sign up for an event, or just find whatever info they're looking for. Right now, it's not quite doing this very well."

To improve the website’s organization and discoverability of these key features, I needed to understand where exactly people were getting stuck and giving up.

I defined my research goals (below) and got to work: Based on the business goals mentioned above, instead of a SWOT competitor analysis, I did a heuristic evaluation of similar sites' homepages and then conducted in-depth user interviews with usability tests with six early-stage business owners, including three existing QEDC users.

Key findings:

To put it lightly, there were...a lot of challenges that users faced on the site while trying to complete their tasks. I wanted to clearly define and identify the biggest confusions and frustrations—but also what's working—to understand what changes would be most impactful. So I mapped out my IDI findings on this affinity map.

Before I made any decisions about improvements and changes to the site, I got clear on who exactly I'm designing for. I developed two personas based on the insights from my IDIs. With my project objectives and goals in mind, I centered the site improvements around the older career changer who is also an ESL speaker—my primary persona. According to Michael, ESL speakers comprise a large share of the QEDC audience, but I heavily relied on both to inform my proposals for improvements because, as it turned out, the two segments shared some specific common ground.

One insight that stuck out to me from the usability testing: I observed that nearly all participants started at the top navigation to find what they were looking for.

So a major piece of making the site easier to use and improving the findability of key features was ensuring that the information architecture is set up in a way that is intuitive and straightforward. The current navigation went as deep as four or five levels, making it difficult for anyone—much less someone who speaks English as a second language—to find what they're looking for. In fact, one out of two ESL speakers I interviewed during usability testing could not find the video library.

I created an affinity map to re-organize the pages based on user needs from my research and edited the sitemap so that the navigation only included the major pages. I then conducted a card sort study to validate my first pass (the post-its below) and realized I still had work to do to. I ended up doing several iterations of the sitemap until the card sort results aligned with my navigation re-work.

I proposed the following changes to the site's IA:

Then, to redesign the homepage's content hierarchy and modernize the UI, I proposed using from Elementor, a free drag-and-drop website builder for WordPress, for two reasons: 1). ease of implementing a redesign, and 2). automatic responsiveness across devices. Click to enlarge the below first drafts!

After receiving stakeholder alignment on my proposed adjustments and hi-fidelity homepage redesign, I conducted two rounds of user testing via Maze to validate the site improvements:

→ First, a tree test to evaluate the findability of features on the homepage to determine any flaws in the updated information architecture.
‍
→ Second, a task-based usability test to validate any revisions needed following the tree test, and to gather behavioral and attitudinal information to measure the success of the homepage redesign.

In the first round of user testing at the beginning of the project, users successfully located the video library under 'Resources,' In this tree test, I was able to test my assumption that changing "Resources" to "Learn" would yield equally successful results. However, only 1 in 10 users clicked on 'Learn' in the tree test to find the videos.

Additionally, out of 11 testers, 7 could not locate the resources for reopening—which is another page I had placed under a menu label called "Learn." Before the next round of usability testing, I changed this label back to "Resources."

The adjustment is reflected in the sitemap below.
‍

Then, using the high-fidelity prototype I created using Figma, I tested the revised design with 10 total users: 3 QEDC users and 7 non-QEDC users.
‍

While the real outcome is yet to be determined (the KPIs and metrics to measure success of the QEDC site redesign—which is currently still in the works—are an increase in visitors landing on the calendar, the page to book appointments with a business advisor, and the video library), this project was a great reminder of one important lesson: Redesigning a site doesn't mean that everything should or needs to change. One of the biggest updates I made to the site's global navigation was invalidated in testing and was ultimately reverted back to its original label—which is, in the end, a happy discovery!

As a four-year resident of Long Island City and a regular at the many diverse restaurants and “mom and pop” shops around the borough from LIC to Jackson Heights to Flushing, I feel particularly drawn to QEDC's mission for supporting and promoting local businesses, and I look forward to how these recommendations can help further the mission of the QEDC!

→ Interact with the final prototype