Rebuilding a homeware store around the product page
A concept rebuild of a mid-size homeware store. The original theme buried the things that sell furniture — scale, materials, and delivery cost — below three screens of brand copy. The rebuild puts them above the fold.
The problem
Vela sells mid-priced furniture and homeware — the kind of purchase people research for a week before committing. The stock theme treated every product like an impulse buy: one image, a price, an add-to-cart button, and then eight hundred words of brand story before any of the information a buyer actually needs.
Three things were missing above the fold on every product page: a sense of physical scale, an honest look at the material, and the delivery cost. Those are the three questions that decide a furniture purchase, and all three required scrolling or a trip to a separate page.
The product page before and after. Scale reference, material detail and delivery cost all move above the fold; brand copy moves below the buy box.
What I did
- Rebuilt the product template in Liquid around a three-part buy box: gallery, scale reference, and a delivery estimate that reads the visitor’s region.
- Designed a reusable material-detail section in Figma — macro photography paired with a plain-language care note — and built it as a theme section editors can reorder.
- Replaced the carousel of lifestyle shots with a fixed gallery, so the images a buyer needs are never behind a swipe.
- Moved brand narrative below the buy box and cut it from roughly 800 words to 120.
- Wired a back-in-stock flow through Klaviyo so sold-out variants capture demand instead of dead-ending.
Targets
- Mobile LCP
- Under 2.0s
- Measured on the concept build; the stock theme rendered at 4.1s on a throttled 4G profile.
- Above-fold answers
- 3 of 3
- Scale, material and delivery cost, previously 0 of 3.
- Product page weight
- −58%
- Measured build-to-build after replacing the carousel and compressing gallery assets.
What I’d do next
The delivery estimate currently reads region from the visitor’s IP, which is fast but wrong often enough to be worth revisiting — a postcode field on the buy box would trade a little friction for a number people can trust.
The material section deserves a video treatment. Fabric and timber both read poorly as stills, and a four-second loop would answer more questions than another photograph.
This is a self-directed concept project. Figures shown are measured against the concept build, not a live storefront.
Working on something like this?
Book a free 30-minute call and we’ll talk through what your store needs.
Start a project