Most agency websites imply that one team did everything. Most good ecommerce builds are not made that way. The design skill that produces a store worthy of a fashion brand and the platform knowledge that makes Shopify Plus do exactly what the design demands rarely live in the same building, and pretending otherwise is how brands end up with beautiful stores that do not work, or working stores that look like everyone else's.
Cult Gaia was built by two studios. ToyFight brought the creative direction, the visual language and their own front-end code. Graftstudio brought the Shopify Plus architecture and the implementation that made the design function on the platform. This is how that split worked in practice, for any agency wondering whether to bring a specialist in.
Who owned what
The clearest way to run a joint build is to make ownership explicit before anything is built. On Cult Gaia it looked like this.
ToyFight owned the design and the front end. The visual system, the layouts, the typography, the interactions and a large part of the component code. Anything a customer sees was theirs to decide.
Graftstudio owned the platform. Theme architecture, how the catalogue was modelled, the integrations for cross-border commerce, personalisation and payments, performance across markets and devices, and the parts of Shopify Plus that determine whether an ambitious design can be shipped without compromise.
The seam between the two was the theme's data and structure. ToyFight designed against it; Graftstudio built it so the design had what it needed and the merchant's team could manage the store afterwards without a developer.
Why the split produced a better store
Cult Gaia's products are high fashion, highly visual and curated. The store had to present them in a way that felt elevated without getting in the way of shopping. That is a design problem and a platform problem at the same time, and the two pull against each other.
A design-led team working alone tends to hit Shopify's edges late, when a layout that looked right in Figma turns out to fight the way collections, variants or markets work. A platform-led team working alone tends to compromise the design early, because it knows the edges and designs around them. Neither produces the store the brand deserves.
With both in the room, the conversation changed. When a design decision had a platform consequence, it was flagged before it was built. When a platform constraint threatened a design, the question was how to keep the design and solve the constraint, rather than which one to give up. The finished store feels like Cult Gaia, which sounds simple and is the hardest thing to achieve for a brand with this much visual identity.
The unglamorous parts
The store handles a wide range of product types without feeling cluttered. Cross-border commerce runs through Global-E, personalisation through Nosto, and payment flexibility through Klarna. None of it is visible as a feature, and all of it had to be integrated so it stayed out of the way of the experience. That is platform work, and it is most of what a Shopify Plus developer spends a joint build doing: making sure the impressive parts have something solid underneath.
The build took sixteen weeks, and the merchant's team can update and manage the store without going back to a developer for every change. That last point is a design requirement as much as a technical one, and it was in scope from the start.
What agencies get wrong about this
The mistake is treating the developer as a supplier who receives finished designs and builds them. That model puts the platform conversation at the end, where it is most expensive. The model that works is the one used here: a specialist who is in the room from the start, whose name goes on the work, and who is introduced to the client rather than hidden behind the agency.
That is also how Graftstudio prefers to work with agencies, whether on a full build like Cult Gaia, on overflow, or as a fractional developer inside the agency's team. The agencies page sets out how that works.
