Every project begins with a base drawing that is already wrong. The building has moved, the drawings are incomplete, the previous survey measured the wrong things. The useful question is not whether the base drawing is accurate. It is whether it is accurate enough for the decision at hand, and which errors matter.
We spend more time than clients expect on survey. Not because we love measurement, but because the survey is where the design argument starts. A plan that records every pipe but misses the fall of the floor is not more accurate than a plan that records the floor and omits the pipe. It is accurate about the wrong subject.
The survey as a set of choices
A measured drawing is a compression. It turns a building into lines, and the lines carry a point of view. The thickness of a wall, the position of a door, the decision to show a ceiling void or a floor finish: each is a choice about what the next person needs to know.
At Kiln Yard, the first survey showed the kiln hovels as solid. They were not. The later fabric survey, done with a conservation engineer, showed voids, repairs, and a brick skin that had been rebuilt after a fire. The line on the first drawing was not a lie. It was a summary. But it would have led us to a different, and worse, structural strategy.
This is why we now survey in layers. The first layer is geometry: enough to plan and section. The second is fabric: materials, condition, and change. The third is use: how people move, where they pause, what they avoid. The three layers are drawn separately and then overlaid. The overlay is usually where the project becomes clear.
What the camera cannot do
Photographs are useful, but they flatten. A photograph of a wall records surface. It does not record thickness, construction, or the way a room sounds. It does not tell you that the floor slopes towards the door because the joists have settled, or that the window is the only source of cross-ventilation.
A section drawing is a better instrument for existing buildings because it shows relationships. It shows the wall meeting the roof, the floor meeting the ground, the new structure meeting the old. Most of the difficult decisions in an adaptive reuse project live at those junctions. A plan can hide them. A section cannot.
We draw sections early, before the design is fixed. They are not presentation drawings. They are questions. What is holding this up? Where does the water go? What happens if we cut here? The drawing becomes a way of asking the building to explain itself.
The ethics of the base
There is an ethical dimension to the base drawing that is rarely discussed. The survey is usually commissioned by the person with the power to change the building. What gets recorded, and what gets left out, shapes what is possible. A tenant’s improvised kitchen might be omitted because it is not part of the historic fabric. A landlord’s blocked door might be shown as a wall because it is not in the lease drawings. The drawing is never neutral.
We try to record use as carefully as fabric. Who works in the cold room? Where does the delivery arrive? Which door is propped open all summer? These observations are not soft data. They are evidence about how the building actually performs, and they often suggest the cheapest and least destructive intervention.
At Northbank, the most important survey finding was not structural. It was that the public path already ran through the site, informally, over a broken kerb and under a fire escape. The consented scheme kept that route and made it legible. The drawing that mattered was a dotted line on a plan, not a measured wall.
Drawing for the next person
A good survey is also a handover document. It should be legible to a contractor, a conservation officer, and a future architect who has never been to the site. That means clear conventions, a stated date, a note on what was not surveyed, and a scale that can be checked. It means resisting the urge to make the drawing beautiful at the expense of being useful.
We still make the drawings beautiful, but beauty is a by-product of clarity. A drawing with consistent line weights, a readable north arrow, and a proper scale bar is easier to trust. The conventions of the architectural sheet are not decoration. They are a contract about what the lines mean.
The existing building will outlast the drawing. The drawing’s job is to give the next person a fair start.
