Photographs that hold up: evidence on site
Most site photographs are taken to jog the photographer's memory. The ones that matter are read months later by somebody who was never there, sometimes by a solicitor, and the gap between those two purposes is where a lot of otherwise good records fail.
The two-shot rule
A close-up of a crack could be any wall in any building. Take two frames of anything that might be argued about: one wide enough to place it — the doorway, the window, the corner of the room — and one close enough to show the defect itself. The wide shot is the one people forget, and the one that makes the close-up admissible in an argument.
Show scale
"A crack" means nothing; a crack against a tape or a coin means something. Anything where size determines severity — cracking, gaps, level differences, standing water — needs something in frame for scale. Use a tape where the number matters and you need to be able to read it later.
Mark it up, and keep the original
An arrow drawn on the image is worth a paragraph of description, particularly on a busy photo where the defect is one of many things visible. But keep the unmarked original too: an annotated image is your interpretation, and being able to produce the untouched frame is what stops that becoming a point of dispute.
Timing decides what a photo proves
A photograph proves the state of something on the day it was taken, and nothing else. That makes the date the most important thing attached to it — and it is why handover photography matters so much. Photographs taken the day you took possession settle the "it was like that when you got it" argument in a way that nothing taken three months later can.
For the same reason, photograph the fix as well as the fault. A closed item with a completion photograph is closed; one marked complete with nothing attached is a claim.
Keep the photo with the item
The most common practical failure is not the photography, it is the filing: four hundred images in a phone roll, in time order, with no reliable mapping back to the defects they belong to. By the time you sit down to write the report, half of them are unidentifiable.
Attach the photograph to the item at the moment you take it, on site. That is the single habit that separates a usable record from a folder of pictures — and it is why Snagger takes photos into the pin rather than into the camera roll, so the evidence and the defect never come apart.