The forty-year test
A building's opening photographs tell you almost nothing. We judge our work by how it looks in its fortieth year — and design backwards from there.
There is a photograph of almost every building in the world taken within a week of its completion: empty, swept, shot at dusk with the interior lights on. It is the least honest image a building will ever produce. Nobody has lived in it. Nothing has failed yet. The sealant is still elastic and the render is still white.
We have a rule in the studio that any material proposed for an external surface must be accompanied by a photograph of the same material, in a comparable climate, at forty years old. Not a manufacturer's render. A photograph of a real building that somebody has had to maintain.
The rule sounds pedantic. In practice it removes about seventy per cent of the palette immediately, and the seventy per cent it removes is almost exactly the portion of the market driven by fashion rather than performance. Composite panels go. Most applied renders go. Almost all sealed joints go, because sealant has a service life of fifteen to twenty years and no residential body corporate in the world has ever budgeted for a full re-seal.
Ask not whether a material looks good on handover day. Ask what it looks like after forty monsoons, and whether anyone will be willing to pay to fix it.
What survives is unglamorous and very old: stone laid dry, concrete cast well, brick, timber that is either detailed to stay dry or accepted as a material that will silver. Metals that patinate on purpose. These are not nostalgic choices. They are the materials whose failure mode is legible and gradual rather than sudden and catastrophic.
The economic argument matters more than the aesthetic one. A building that requires a major envelope intervention at year twenty transfers a cost to a future owner who had no say in the specification. That is not a design decision. It is a debt, and it should be described as one.
The forty-year test does not make buildings beautiful. It makes them survivable, which turns out to be a precondition for anything else.
Written by
Ira Naravane
Published 14 July 2026. Aurelis publishes the reasoning behind its decisions, including the assumptions. Corrections and challenges are welcome at studio@aurelis.example.
We don’t just build properties. We create landmarks.
We take on four projects at a time. If you are considering a site, a collaboration or a commission, the conversation starts here — and it starts with a real person reading it.