Proof · Case studies · Study

Luma to Hyvä, on the platform they already owned.

Half Time Beverage: deep catalog, real customization, frontend performance costing conversions. Solved with a frontend rebuild, not a replatform.

Study · Half Time Beverage

The problem lived in the theme layer, so that’s what we replaced.

Context

Beverage retailer, deep catalog, established Magento store carrying real customization. They had to choose whether to stay or go. Subscriptions made staying on Magento the clear choice.

Constraint

Frontend performance was costing conversions, but the backend investment was sound. A replatform would have thrown out years of working integration for a problem that lived in the theme layer.

What we did

Rebuilt the frontend on Hyvä: Tailwind-based theme, Alpine.js interactivity, checkout preserved, custom catalog features reimplemented natively, not forced.

Chrome UX Report LCP history for halftimebeverage.com: 52% good in July 2023,
88% good from August 2023 onward, p75 1.7 seconds
Google’s Chrome UX Report for halftimebeverage.com: the Hyvä launch is the line between July and August 2023.
6.5s → 1.9s
median page load, the night of the Hyvä cutover (RUM, Jul 30 2023)
 
52% → 88%
page loads with good LCP, Jul → Aug 2023 (CrUX); p75 1.7s, still green on every Core Web Vital today
 
+147%
organic traffic, Jul → Aug 2023 (Analytics)
 
RUM page-view load time chart, July 27 to August 2 2023: median around 6.5
seconds drops to under 2 seconds at the July 30 cutover
The cutover night in RUM: page-view load time, Jul 27 – Aug 2, 2023. You can tell where it changed.

Engagement type: project, delivered under an ongoing retainer.

Same class of problem on your store?