Case study · Retail / Furniture
"View in your room" WebAR for a furniture retailer
Tap a button on any product page, place the sofa in your living room through the phone camera, then check out, no app install, no friction. Returns dropped. Confidence went up.
The challenge
What we had to solve.
Furniture is a high-return category for a reason, the customer can't tell if the couch is the right size or shade until it's in their living room, and by then it's already a return.
Native AR apps had been tried at this retailer and abandoned: customers didn't install, the catalog wasn't worth a download, and the app got buried on screen 3.
Our approach
How we tackled it.
We built it for the web, not as an app. The WebAR flow lives on the product page: a single button, native AR on iOS and Android via the browser, no install ever.
We modelled the top 200 SKUs (the long tail by revenue) at production-grade quality, with accurate materials, dimensions and shadow casting. Cheap models look like toys; we didn't ship cheap models.
We tied the AR view to the cart: once a customer placed a sofa in their living room and got the "yes" feeling, the "add to cart" button was one tap away, with the right colour, size and configuration carried over.
We A/B tested AR-enabled vs. catalog-only product pages for three months, measuring return rate, conversion rate and time-on-page.
Deliverables
What we built.
Specific, named outputs, not vague "strategy".
WebAR experienceBrowser-based AR on iOS + Android, no app, accessible from product page.
3D model catalogTop 200 SKUs modelled to production quality with accurate materials and dimensions.
Cart integrationConfiguration carries from AR view to checkout.
A/B testing frameworkThree-month controlled comparison vs. catalog-only product pages.
Performance dashboardAR engagement rate, AR-to-cart, AR-to-return-rate delta by SKU.
Outcomes
What it returned.
- Return rate on AR-previewed items 27% lower than catalog-only-purchased items.
- AR-enabled product pages converted 1.6x catalog-only pages on visitors who interacted with the AR button.
- Customer support tickets about dimensions dropped meaningfully.
- Three sister retailers in adjacent categories asked for the same build, the system was designed to extend.
The takeaway
What we learned.
WebAR works because the friction is gone. Customers don't "decide to use AR", they tap a button. The win wasn't the technology, it was the placement: native AR on a product page is just a better product page.
Services used
How we got it done.
Related work