Introduction
Safety reviews are often framed as a search for more information: more photographs, more measurements, more notes. Many reviews already have plenty of fragments. What they lack is the spatial relationship that makes those fragments interpretable.
This article is about context versus data. It is not about whether an investigator collected enough numbers. It is about whether later reviewers can still see how conditions related to one another in the environment.
Data answers isolated questions. Context explains relationships
A photograph can show a surface clearly. It does not, by itself, show distance to an access path, elevation relative to equipment, or what sat next to it. A note may describe a condition accurately and still omit how people moved through the space.
Context includes orientation, scale, adjacency, and environment. Without it, later reviewers interpret fragments in isolation. That is a different failure mode from “we needed one more measurement.”
Context, in this sense, is a shared spatial frame. It includes the direction a camera faced, the distance between a condition and a path or piece of equipment, the elevation of a work surface, and what occupied the surrounding area at the time of the visit. A single measurement can be correct and still leave those relationships unrecorded. Reality capture—photographs or scans organized so a later reader can move through the place rather than through a folder of separate frames—is one way to keep the frame with the detail. The capture records the scene as it was found. It does not decide which relationship a later review should treat as relevant.
Why more fragments still leave a gap
Photographs, inspection notes, and measurements are useful. They are usually collected from convenient viewpoints, written against what seemed relevant at the time, and stored without a shared spatial frame. Each piece can be locally accurate while the relationships between them are not preserved.
The gap is easiest to see by asking what a later reviewer can still locate. A close-up can confirm the appearance of a surface. It cannot, by itself, show how far that surface sat from a doorway, a ladder, or a machine unless the capture was planned to preserve those relationships. Photogrammetry derives shape and position from overlapping photographs and can produce a mesh, an orthomosaic, or a point cloud. A point cloud is a set of measured points describing surfaces in space. An orthomosaic is a plan-view photo map assembled from those overlapping images. An ordinary set of photographs does not become any of those products because the files share a folder. Each product is also limited by overlap, control, and processing. None of them is a property survey.
The practical result is often a large set of locally accurate records that cannot be placed. Several close-ups of the same guard, stain, or cover may still leave a reviewer unable to say which wall, which level, or which side of a machine they show. A note that says “north wall” or “near the press” depends on a plan or on someone’s memory of the room. After the room is rearranged, the words and the photographs no longer point at the same place. Existing-conditions documentation is a dated record of the site as it was found, kept so those pointers remain available. It preserves relationships. It does not rank which fragment later mattered.
What missing context forces later reviewers to do
When that dated frame was never captured, later review does not stop. It changes the kind of work reviewers have to do.
They reconstruct distances, access, and proximity from memory or partial records. Relevance becomes subjective. Stakeholders can disagree about whether nearby conditions mattered. Human-factors writing, including Reason’s work on latent conditions in systems, treats layout and interaction as part of how events emerge. NIOSH Total Worker Health similarly treats work as a system rather than a set of isolated observations. Those are frameworks for thinking about systems, not methods SterFlies applies as an investigator.
Isolated photographs and notes have their own limits. That is the subject of photos, notes, and memory as investigation records. The timing problem—review starting after the scene changes—is covered in why documentation is often requested too late.
Forensic site documentation is one way to keep those relationships available for later qualified review. It does not make a record “defensible” in a legal sense, and it does not determine root cause.
Conclusion
Safety reviews run into this limit when the file can describe pieces of a site and cannot show how those pieces sat together. Additional photographs and notes can leave the same limit in place if they were never tied to a shared view of the environment.
A spatial record gives later reviewers one reference for orientation, scale, and adjacency. An interior walkthrough, an orthomosaic, or a point cloud can each do part of that work, within the limits of how it was captured and processed. Reviewers can then ask where a condition sat, what surrounded it, and which path or piece of equipment was nearby, using the record rather than a reconstruction from memory.
SterFlies documents the site so qualified reviewers can still see those relationships after conditions change. The record is a reference for that review. What the relationships mean for an investigation remains with the people responsible for the review.