We photograph buildings that do not exist yet

Most visualisation fails in the same way: it is too clean. Nothing is worn, the light comes from nowhere in particular, and there is no evidence a person has ever stood in the room. The image looks expensive and says nothing.

We work the other way round. We start from the argument the building is making — about its site, its material, the way you enter it — and we build the frame that makes that argument impossible to miss. Then we age the concrete, throw dust into the air, and put someone in the frame who is late for something.

It means we ask uncomfortable questions early. It also means the committee usually says yes.