One of the major projects I spearheaded at Borrowed & Blue was a complete site redesign that focused on modernizing/simplifying our codebase and improving the user experience. I worked extremely closely with our lead Rails engineer to turn this complicated redesign around on a short time frame.
The existing data models and information architecture forced the site into an extremely siloed organization based on "areas". This organization had been useful as Borrowed & Blue slowly launched in specific areas, but at scale, they were a huge headache. Each vendor and wedding was tied directly to a specific "area" and one area only.
Because the methodology around creating an area wasn't well defined, it was possible for vendors who had a "home base" in Boulder, Colorado to not be listed in Denver, Colorado – merely 20 minutes away, because Boulder and Denver were distinct "areas".
This kind of organization was a decent strategy for local SEO, but significantly hampered growth and usability once B&B decided to launch "nationwide" with their vendor and wedding database. The database models hadn't been designed with flexibility in mind and the entire product was essentially locked into these fixed silos. Our architecture was siloed in a [location]/[category]/[profile] structure.
I designed a new paradigm for the site that removed our reliance on areas, and towards a faster, simpler search method based on locations + radius. The design and front-end architecture was arguably the easy part – the back-end work required significant changes to almost every model in our database. I worked closely with our head of engineering to figure out how to move us forward in a flexible and performant way.
This also involved some repositioning on the vendor side, and we built a new version of our vendor dashboard to focus on profile completeness and new features.
While the site sported a nifty new design and simpler UX that improved our site usage and laid the groundwork for several new feature directions, perhaps the most significant improvement was in our codebase. We updated the app 2 full Rails versions, cut our front-end code and CSS overhead significantly, and smarter organization and usage of data led to nearly 3x increases in page speed.