Xumulus Logo
Search
Agent-ready search & discovery on Next.js + Elastic/OpenSearch

Commerce website design and build

We design storefronts around the catalog behind them. On a large catalog the highest-traffic templates are search results and category pages, and those are the ones a portfolio-led design process tends to leave until last.

Elastic
Relevance + speed
Next.js
Modern UX + performance
RAG
Agentic discovery ready
AI Discovery Signals
Retrieval-first architecture makes agentic discovery possible.
Scroll

Design that survives contact with the catalog

A homepage is comparatively easy. The pages that decide revenue are the results page, the facet rail and the product template, and they have to hold up against products with missing images, forty-character option names and attributes that only half the catalog has.

So we design against a real export rather than placeholder products, and we build on the platform you run — BigCommerce, Shopify Plus, Adobe Commerce, or a Next.js front end over an Elastic or OpenSearch index.

Discovery bottlenecks

What we design for that comps usually skip•

These are the states that make up most sessions on a large catalog.

01
The results page as a primary template
On a catalog of any size, search and browse carry more traffic than the homepage. Facet layout, result density, sort affordances and empty states are the design work that moves conversion, and they are rarely in the comp.
02
Products with incomplete data
Missing images, absent attributes and inconsistent option naming are the normal condition of a real catalog. Templates have to degrade cleanly instead of rendering an empty container or the literal words "no image".
03
Mobile facet interaction
Filtering is the hardest interaction on a commerce site and the one most often inherited from the theme unchanged. It deserves the same attention as the checkout.
04
Performance as a design constraint
Layout decisions set the image payload and the interaction cost before a single line of code is written. Treating speed as an optimisation phase after design is how sites end up slow by construction.

Have a storefront that needs rebuilding?

Send us the site and roughly what the catalog looks like. We will tell you whether this is a design problem, a data problem or a search problem — they are easy to mistake for each other.

Search is the foundation of AI commerce