Branded merchandise stores were being built one at a time — the same blank products re-entered by hand, the same artwork re-placed, and a mockup produced manually for every colour and size. We built a platform that generates a store, connects it to a supplier catalogue, and applies artwork to the selected blanks, rendering a preview for every variant automatically.
A branded merchandise store looks like a small e-commerce site and behaves like a manufacturing configurator. The products are not owned: they come from a supplier's catalogue of blanks, with their own colours, sizes, stock levels and prices that keep moving. The thing being sold is not the blank but the blank plus a print, and what the customer must see before buying is that combination — a specific artwork, on a specific colour, in the right place, at the right size. Doing this by hand meant re-entering the same hoodies and t-shirts for every new store, then producing a mockup image per colour per print, which is a designer's afternoon multiplied by everything in the shop, and which goes stale the moment the artwork or the supplier's colour range changes. On top of that, each product type constrains what can physically be printed on it — print area, technique, maximum size, resolution — and a store that ignores those constraints simply forwards impossible orders to production.
A new store is generated rather than built: domain, theme, brand colours, logo, pages and legal text are configured and the storefront exists. Several stores run on the same engine and the same catalogue while keeping separate branding and separate front ends, so the second store costs a fraction of the first.
Products come from the supplier catalogue with their variants — colours, sizes, stock and prices — rather than being typed in. The operator picks a selection out of the catalogue and it becomes the store's assortment; subsequent syncs keep prices and availability current instead of leaving a store selling blanks that were discontinued.
Artwork is uploaded once and positioned on a product's print area, with placement and scale stored against the product type rather than against a single image. Moving a print or swapping the artwork re-applies everywhere it is used, which is what makes a catalogue of hundreds of decorated items maintainable by one person.
The artwork is composited onto the product image for each colour of each selected blank, so the storefront shows the actual combination a customer is buying without a photo shoot or a designer producing images by hand. New colour in the catalogue, new preview.
Each product type carries its print area in millimetres, the available decoration techniques and a minimum resolution. Artwork that would print at too low a resolution or fall outside the area is rejected at upload with the reason, and the decoration cost is added to the item price by technique and size — so what the store sells is what production can actually make.
An order line is not just a SKU and a size: it carries the artwork file, the placement, the technique and the print area it was validated against. Production receives a specification rather than a screenshot and an email thread, which is where most of the manual reconciliation in custom merchandise usually happens.
Launching a store became a configuration task rather than a project, and decorating one became an upload rather than a design job repeated per colour. Because assortment comes from the catalogue and previews are generated per variant, a store's product range stays consistent with what the supplier actually has and what production can actually print. The platform has been in production use for several years across more than one store.
Tell us what the process looks like today and we will tell you what can be automated — and what should not be.
LET'S TALK
