Bubbly BelleDTC E-commerce5 min read
Nobody handed Bubbly Belle a design system. There was no brand guide, no component library, no prior designer to hand things off. There was a product, bath bombs and beauty bundles, and a bet that the right brand and buying experience could turn it into something real. I was the only designer. I owned everything from the logo to the checkout confirmation screen. What followed was 18 months of building, testing, and compounding small wins into a business doing over a million dollars a month.
01 / Strategy
No one buys from a brand they don't believe in. Before touching the product pages or checkout, I built a visual identity from scratch, logo, color palette, typography, photography art direction, packaging. The goal was to position Bubbly Belle as a premium but approachable gift brand. Something you'd feel good giving. That foundation meant every downstream design decision had something real to build on.
I studied the decision loop for beauty and gift products, what creates hesitation, what creates desire, what makes someone add to cart versus leave. The answer was trust signals. Clear ingredients above the fold, social proof in the buying decision zone, lifestyle imagery that let the product feel like an object you'd actually use. I restructured the page hierarchy to put the emotional case for the product first and the transactional mechanics second. That reorder alone moved the conversion rate.
The original checkout had drop-off points I could measure and explain. I collapsed it into a single-page flow with progressive disclosure, address, shipping, and payment in one view, with a persistent order summary so users never lost their place. On mobile it stacked cleanly. The number of people who started checkout and didn't finish dropped significantly. You don't need to be clever to fix checkout. You need to stop making people work.
I ran A/B tests on every major surface, hero imagery, CTA copy, button placement, bundle presentation, urgency signals. The 117% CVR lift wasn't a single win. It was a dozen decisions that each moved the conversion rate a few points in the right direction, compounding over time. I kept a running log of every test: hypothesis, variant, result, and what we did next. That discipline is how small teams move fast without making the same mistake twice.
Acquisition gets the credit, but retention is where DTC brands make their margin. I built a library of email templates, welcome series, post-purchase, winback, seasonal, designed to the same standard as the site. Consistent brand, clear hierarchy, mobile-first. They didn't feel like blasts. They felt like the brand showing up.
02 / Results
03 / My role
I was the only designer at Bubbly Belle for the full run. That meant complete ownership and zero places to hide. I built the brand identity from scratch, designed and iterated every customer-facing surface, directed product photography, wrote the design specs that handed to developers, and ran the testing program that generated the results. I worked directly with the founders, which meant fast decisions and tight feedback loops. There was no committee. There was just the work, the data, and the next test.
My psychology background informed how I thought about the conversion problem, not as a layout question, but as a trust and motivation question. What does this person need to believe before they buy? What's creating hesitation? What resolves it? That frame produced better hypotheses than starting from the UI.