Skip to content

Case study · AI + SaaS Products

A portfolio the studio owns outright

The practice did not need a showy site. They needed one where the buildings carry the page, and that they own outright.

What most agencies push
  • Something showy, heavy with interface
  • Built so changes route back through the agency
  • Clients notice the site, not the work
Client
an architecture practice
Problem
A portfolio risked showing off its own interface instead of the work.
Approach
Strip the interface back until the projects carry the page, then hand it over.

The problem

The practice came to Stallwart for a portfolio site. The default a studio is usually offered is something showy, heavy with interface, and built so that future changes route back through the agency that made it.

That serves the agency more than the studio. The work is architecture, so the real question was whether the site could get out of its own way and let the buildings carry the page.

The reality behind the problem

The showy default is attractive because it looks like value: more interface reads as more work delivered. But on a portfolio, every element the eye lands on is an element that is not the architecture, and a site that performs its own cleverness quietly competes with the buildings it exists to show. The question was never how much to add. It was how little the page could carry and still do its job.

The second cost is slower to show and harder to undo. A site built so that changes route back through the agency keeps the studio tied to whoever made it. That dependency is easy to sell and expensive to live with, because every future edit, however small, becomes a request rather than something the practice can do itself. A portfolio the studio cannot change on its own is not fully theirs.

What was assumed

A portfolio should look impressive.

What Stallwart asked

What if the work, not the interface, carried the page?

The desired outcome

The site had to disappear behind the work and stay fully in the practice's hands, long after Stallwart stepped away.

  • The projects carry the page, with nothing in the interface competing for attention.
  • Pages load fast enough that a visitor never waits to see a building.
  • The practice can change the site themselves, with no standing dependency on Stallwart.
  • The design stays restrained by intent, not plain by accident.
  • The whole site, code and content, handed over for the studio to own outright.

The system we designed

Stallwart stripped the interface back until the projects carried the page, built it to be fast, and handed it over for the practice to own outright, with no dependency on Stallwart to change it later. Clients open it and talk about the buildings, not the buttons.

How it works

  1. 01

    Decide what the page is for

    The page exists to show the work. Everything that competes with the architecture for attention is a candidate to remove, starting from that.

  2. 02

    Strip the interface back

    Remove interface until the projects carry the page on their own. What stays is what the work needs, not what makes the site look impressive.

  3. 03

    Build it fast and hand it over

    Deliver something fast that the studio owns outright, with no standing dependency on Stallwart to make a change later.

The approach

Less interface, so the work carries the page.

What most agencies push

  • Something showy, heavy with interface
  • Built so changes route back through the agency
  • Clients notice the site, not the work

What Stallwart handed over

  • Interface stripped back until the projects carry the page
  • Fast, and owned outright
  • Clients open it and talk about the buildings

Architecture

What sits underneath.

The site was built in plain layers the practice can hold in their head, each chosen so the work stays first and the studio stays in control.

01

Content and structure

The projects are the content, organized so a visitor moves through the work directly. The structure is simple enough that the practice can add or reorder a project without calling anyone.

02

Rendering and interface

The interface is stripped back to what the work needs. Pages are built to present buildings cleanly rather than to perform effects, so nothing on screen draws attention away from the architecture.

03

Performance

Images and pages are built to load fast, because a portfolio that makes a visitor wait has already lost the attention it was meant to hold on the work.

04

Hosting and ownership

The site is hosted and handed over so the practice owns the whole of it, code and content, with no part of it locked behind Stallwart.

The hard parts

Where the engineering judgment was.

01

Less interface, not more

On a portfolio, interface competes with the work. Removing it lets the buildings carry the page.

The trade-off

A restrained site looks less busy than a showy one, which is the intent, not a shortfall.

02

Hand over full ownership

A studio should not depend on its agency to change its own site. Ownership is the deliverable, not a retainer.

The trade-off

No standing contract to lean on, by design. The studio can change the site itself.

03

Fast over elaborate

A portfolio that loads fast respects the visitor and keeps attention on the work.

The trade-off

Fewer heavy effects, which is what keeps it fast and out of the work's way.

Reliability & guardrails

How it avoids the wrong call.

A portfolio has to be dependable the moment a client opens it, which for a small site means staying fast, staying up, and failing quietly if anything does go wrong.

Stays up without tending

The site is built to run without ongoing intervention, so it is there whenever a client opens the link, not only when someone has recently checked on it.

Fast on an ordinary connection

Pages are kept light so they load quickly on a normal device and network, rather than assuming the visitor has the fastest of either.

Degrades gracefully

If an image is slow or a feature is unsupported, the page still shows the work rather than breaking, because the content matters more than the effect.

Simple enough to stay reliable

A smaller site has fewer moving parts to fail. Removing interface removed whole categories of thing that could go wrong with it.

Productionization

What turns it from a demo into software.

What turned the build into a site the practice could run on their own, not a demo that needed Stallwart nearby.

Editable content

Projects can be added, changed, or reordered by the practice directly, so keeping the portfolio current does not route back through an agency.

Documentation

The site comes with plain notes on how it is built and how to change it, written for the studio rather than for another developer.

Deployment the studio controls

Publishing a change is a step the practice can take themselves, so an update goes live without waiting on Stallwart.

A clean handover

The whole site was handed over in a state the studio can pick up, with no part of it left depending on Stallwart to keep running.

What changed

The practice got a site where the buildings carry the page. The interface stepped back far enough that a visitor's attention goes to the architecture rather than to the design around it, which is what a portfolio is for.

It is also theirs in full. The studio can change the site themselves, without a standing line back to Stallwart, so the portfolio stays current as the practice's own task rather than an agency's recurring invoice.

  • From a site that shows off its interface to one where the work carries the page.
  • From changes that route through an agency to edits the studio makes itself.
  • From a portfolio that performs to one that gets out of the way.
  • From a standing dependency to full ownership.

In practice

Principal Architect, an architecture practice

The practice's principal architect described Stallwart stripping the interface back until the projects carried the page, delivering something fast that the studio owns outright, and walking away rather than leaving him dependent on the agency to make changes. Clients now open the site and talk about the buildings, not the buttons.

Ownership & handover

What you receive, and keep.

The site is the practice's to run and change, with nothing held back.

  • Source code for the whole site.
  • The content and images, in a form the practice can edit directly.
  • Hosting and deployment set up in the studio's own hands.
  • Documentation of how the site is built and how to change it.
  • No standing dependency on Stallwart to make an update.

What we learned

The insight that generalizes: on a portfolio, restraint is the work. The hardest part is not deciding what to add but having the discipline to remove interface until the thing the page exists for carries it alone. And a site is only truly the client's when they can change it without calling the people who built it, so ownership is a design decision, not a line in a contract.

This work connects to

See it in action in
  • AI + SaaS Products
  • Portfolio and web
  • Full ownership

Frequently asked

What makes a good architecture portfolio website?

One where the work carries the page. For an architecture studio that means a fast, minimal site with the interface stripped back, so visitors look at the buildings rather than the site's own design.

Why strip the interface back instead of adding features?

Because on a portfolio the interface competes with the work for attention. Removing it lets the projects carry the page, which is what a visitor came to see.

What does owning the site outright mean?

The practice can change the site themselves, with no standing dependency on Stallwart. Ownership was the deliverable, not a retainer, so the studio is not tied to its agency to make an update.

Does a minimal site mean a plain one?

No. Minimal here means the interface gets out of the way so the architecture carries the page. The result is restrained by intent, with the work, not the buttons, drawing the eye.

Want a site clients look through, not at?

We strip the interface back until the work carries the page, then hand it over yours.

Last updated: October 8, 2026