Design Work.

The site was running on a legacy CMS that had become a bottleneck for everything: editorial work crawled, content was duplicated across pages, and UX improvements kept dying in the backlog. I started with an audit of the existing experience to understand what the site was actually doing for customers versus what the business assumed it was doing. That audit shaped the IA, content, and research strategy for the whole program.

I built personas grounded in real customer behavior across Southwest Gas's three-state service territory: Arizona, Nevada, and California. Each persona was paired with an end-to-end journey covering the tasks that mattered most: starting service, paying a bill, reporting a leak, applying for energy assistance. Those journeys became the shared reference for everything that followed. IA decisions, content priorities, and visual hierarchy all traced back to them.

The existing navigation was organized around the company's org chart, not around what customers came to do. I restructured it around the top customer tasks: bill pay, service requests, account management, and emergency information. Energy savings programs and rebate content got promoted into the navigation rather than buried in marketing pages, because customers were making decisions about equipment and usage that affected their bills every month.

Rather than running strategy, design, and content in sequence, I structured the program as three concurrent tracks: UX strategy and research, design and prototyping, and content and art direction. Each track cycled through a Think-Make-Check loop every sprint. Findings from usability testing fed iterative changes to templates and IA in the same week, not at the end of the project. I ran four rounds of moderated testing with real customers before any designs were considered final.

Before settling on a tone, I ran a brand perception study to understand how customers actually felt about the product they were buying. The study used an experimental design: respondents were randomly assigned to one of four conditions and asked to rate their feelings about one of the following terms: "natural gas," "natural methane gas," "methane," or "methane gas." The goal was to measure how much the name itself shaped perception, independent of any site content.

The finding was clear. "Natural gas" produced strongly positive emotional responses, 76% favorable, higher than oil (51%) or coal (39%). The term "methane" produced the opposite: negative associations with pollution and environmental harm. Respondents connected "natural gas" with words like clean, efficient, and reliable. "Methane gas" connected to pollution and risk. Same product, opposite reactions depending on what you called it.

This told me that the language Southwest Gas used across the site was not a copy decision, it was a design decision. Every label, heading, and CTA either reinforced or worked against a perception the company had already earned. The redesign needed to treat the words as part of the experience, not an afterthought.

The research also surfaced a second, more specific concern: consumers associate natural gas with explosion risk. Not as a daily anxiety, but as a latent safety concern that sits just below the surface. Decades of public safety campaigns have trained people to connect the smell of mercaptan, the sulfur-like additive in natural gas, directly with a leak and immediate danger. When customers smell it, their first thought is evacuation, not appliance trouble.

That finding had direct IA and design implications. Emergency content needed to be in the navigation and findable in seconds, not buried. Safety pages needed to be scannable and action-oriented, with clear instructions that matched the mental model customers already had. And the overall tone of the site needed to be reassuring without being dismissive: warm and credible, not corporate or evasive. You cannot use a soft marketing voice on a page someone lands on at midnight because they smell rotten eggs.

With the perception data in hand, the tone direction became concrete: warm, reassuring, and grounded in the language customers already responded well to. The site needed to feel like a trustworthy neighbor, not a corporate utility. I developed a mood board to align the team on that direction before any screens were designed. Once the team committed to the tone, it governed every copy and design decision that followed, from headline structure to the way error messages were written.

Lead UI designer on both pages. I owned the end-to-end design from wireframe through high-fidelity, collaborated with the product and marketing teams on content hierarchy, and presented design rationale to the VCA stakeholders. The work shown here reflects my individual contribution.

VCA operates hundreds of animal hospitals across the country. Their website needed to serve two distinct moments: someone who has never been to VCA and is deciding whether to try them, and someone who already trusts the brand and just needs to find their nearest location quickly. Those are different jobs with different emotional states, and the design had to reflect that. I started each page with low-fidelity wireframes to settle structure and hierarchy before any visual decisions were made, then moved to high-fidelity once stakeholders aligned on the layout.

The Free First Exam offer is VCA's primary acquisition tool for new pet owners. The people most likely to convert are also the most uncertain: new pet owners who haven't established a vet relationship yet. The design needed to reduce friction, answer the questions that create hesitation (what's included, what to expect, is this the right hospital for my pet), and make the call to action feel like the obvious next step rather than a commitment. I structured the page so that trust signals and FAQs appeared before the form, not after it, based on what we knew caused drop-off on the original page.

The hospital finder had to do more than show locations on a map. Pet owners searching for a vet are often filtering by more than distance: specialties, hours, whether the hospital handles emergencies, and whether they've seen reviews. The design surfaced those filters without burying the list, and kept the map and the results in view together so users could orient spatially while comparing options side by side. The biggest structural decision was keeping both panels visible simultaneously rather than toggling between map and list views, which testing on similar healthcare finder patterns had consistently shown reduces task completion time.

Stakeholder review confirmed that the redesigned Free First Exam page addressed the primary friction points the team had identified in the original: unclear offer scope and a form that appeared before the user had enough context to feel confident filling it out. The hospital finder's dual-panel layout was validated as a structural improvement over the original single-view experience. Both designs moved into development handoff.

SDG&E was not a UI project that happened to include some research. It was an end-to-end UX engagement run by a two-person team: me and one other UI designer. Together we owned the work from discovery through delivery. Research, requirements, IA, UX and UI design, usability testing, accessibility, and executive presentation all ran through us.

I led the research and requirements phase rather than receiving a spec and producing screens. That meant user interviews to understand how customers actually used the product, surveys and an analytics audit to identify where they were dropping off, and card sorting to map the mental models they brought to account management. The existing IA was organized around SDG&E's internal departments, not around what customers came to do. Research made that visible and gave us a foundation to restructure around customer goals rather than organizational ones.

Before we could test whether a design worked, we needed to understand what problem it was actually solving. That's why we ran a focus group before any usability testing. The focus group was generative: eight residential customers who had both gas and electric service, all responsible for paying the household utility bill. The goal was to understand how they thought about energy, not evaluate a design. Usability testing came later, once we had something to test, and answered a different question: does this interface work?

Five themes came out of the sessions. Customers wanted explanation, not just data. The question they were really asking when they looked at their bill was not "how much did I use?" but "why is it this high?" They thought about gas and electric as one household expense, even though the bill separated them. High bills triggered investigation, which meant SDG&E had an opportunity to explain unusual changes proactively rather than making customers hunt for answers. Customers wanted actionable recommendations, with enough context to trust them. And trust itself was the constraint: recommendations landed when customers could see the reasoning behind them, not just the conclusion.

Those five themes became the design criteria. The bill redesign prioritized explanation over reporting. The energy dashboard unified gas and electric into a single household view. The recommendation UI led with what to do and why, not just what the data showed.

Research findings shaped every design decision. Pain points from interviews drove flow prioritization. Card sort results restructured the navigation. Usability test feedback at wireframe stage caught structural problems before visual design locked them in. The process moved from evidence to decision at each stage, rather than from stakeholder preference to design.

I designed mobile-first, starting with the flows that generated the most friction: bill pay and outage information. Bill pay consolidated a multi-page flow into a single scrollable screen with a persistent payment summary. The energy dashboard gave customers a clear view of usage trends without requiring them to interpret raw data. The outage center prioritized status and affected area above everything else, because customers landing there are already stressed and need answers fast.

For the payment flows inside the logged-in MyAccount area, we ran several rounds of Design Thinking workshops with stakeholders and customers. These weren't presentation sessions. They were working sessions where participants mapped their own understanding of the payment process, identified where it broke down, and helped us pressure-test early concepts before any screens were built. The payment flow is where most customer frustration concentrated, and the workshops gave us a shared understanding of why that was before we started solving it.

Usability testing ran at wireframe fidelity first, which kept feedback focused on structure and content hierarchy rather than aesthetics. Testing at hi-fi stage confirmed that improvements held under realistic conditions. Results from both rounds produced specific, actionable findings rather than general impressions.

Accessibility testing against WCAG 2.1 AA ran in parallel with design rather than after it. That meant findings could shape decisions rather than just document failures at the end of the project.

The work went in front of the CEO, which required translating UX findings and design rationale into business language: connecting research insights to outcomes an executive audience would act on. I was also involved before delivery. I helped pitch the initial engagement, which meant explaining the value of a research-led UX process to a client who was still deciding whether to buy it.

SDG&E reinforced something I've come to believe about UX at its best: it's a decision-making process, not a production process. Research establishes what the problem actually is. Requirements define what needs to be solved. Design explores solutions. Testing provides evidence about whether the solution is moving in the right direction. When those steps are connected and running on the same evidence base, the work produces something you can defend. When they're separate, design becomes opinion.

Six designers on this engagement, and I was leading all of them. Five had come from business analyst backgrounds and were making the transition into UX design. Only one had formal UX/UI education. For most of them, this was their first time working through a structured design process from discovery through delivery. My job was two things running in parallel: get the work done, and build their practice while we were doing it.

I set the strategic direction, defined the design approach, and divided the scope by flow. Each designer owned a section of the experience. I took a set of screens myself while the team worked in parallel. That meant I wasn't just directing from a distance. I was in the files, making the same kinds of decisions I was asking them to make, which made the critique sessions more honest.

We ran regular alignment sessions where I reviewed work in progress and pushed the team to defend their decisions. Not just show screens. The sessions were structured as critiques: here's what you decided, here's the constraint you were working against, here's what the design needs to communicate. For a team coming from BA work, that framing was new. BAs are trained to document decisions, not design them. Learning to explain why a layout choice serves the user took practice, and the sessions were where that practice happened.

The BA background was actually an advantage in some ways. They understood the business rules, the edge cases, and the operational constraints better than most designers would going in. The gap was methodology: how to frame a problem visually, how to handle competing constraints in a layout, how to give and receive critique without making it personal. By the end of the engagement, the team was running critique sessions themselves. That's the outcome I was most proud of.

The current MazdaUSA.com homepage leads with a rotating banner and a shopping tool that pushes users to select a model before they've had a chance to feel the brand. The Build & Price experience is functional but cold. Nothing about the site communicates that Mazda is a brand built around driving pleasure and craft. It reads like a transaction, not an invitation.

I started with a full IA audit of the existing site: 60+ nodes across four primary navigation sections, mapping where content was buried, duplicated, or missing entirely. From there, I explored a new visual direction that treats the vehicle as the hero. Large-format photography, a restrained typographic system, and a dark editorial palette that signals premium without mimicking luxury competitors directly.

The redesign concept pulls the vehicle forward on every page, reduces chrome, and gives Mazda's design language room to breathe. The mobile experience was a priority from the start: simplified navigation, full-bleed imagery, and a Build & Price flow that doesn't require a desktop to use comfortably.