Racking documentation has a particular life cycle. It is handed over, filed somewhere sensible, and then not looked at by anybody for four years. The person who filed it leaves. The layout changes twice. And then, in a single week, three different people want to see it.

It is never a good week. The requests cluster around events — a lease renewal, a client audit, an incident, a plan to extend. And the answer "we must have it somewhere" carries no weight in any of those conversations.

What the file is actually for

Strip away the folders and a racking file exists to answer three questions:

1. What is this rated to carry, in the configuration it is standing in?
2. What was actually installed, and where?
3. When was it last looked at, and what was found?

Everything worth keeping is evidence for one of those three. Everything that is not evidence for one of them is filing for its own sake.

The four documents

1. The load information, and the notices that summarise it

The safe working load of a run depends on the upright section, the beam profile, the bay dimensions and the configuration as built. The load notice fitted to the frame is the short public version of that — the number a forklift driver can act on. It is not the whole record, and it is only true for the configuration it was written for. Our note on load notices and SWL signage covers what they should say and where they belong.

2. The as-built layout

Not the layout that was quoted — the one that was installed. Run positions, bay dimensions, beam levels, aisle widths. These diverge more often than people expect, because the last bay of a run meets a column, a door swings the wrong way, or a sprinkler drop lands where a frame was drawn. The as-built is what any future work starts from, which is why we keep ours for a long time; why we hold as-built drawings for ten years explains how clients actually use them.

3. The handover file from the installer

Components used, anchoring, and what was handed over on the day. This is the document that establishes what the system is, and it is the one most often never requested in the first place — which is a good reason to know what your installer should be giving you before the crew drives away.

4. The inspection history

When the system was inspected, what was found, and what was done about it. This is the only one of the four that is supposed to grow, and the one that most clearly demonstrates that the racking is being managed rather than merely owned. A single dated record showing damage found and repaired says more than a clean file with nothing in it.

The fifth thing, which is where files go wrong

Modifications. Every beam level moved after installation changes the assumptions the rating was built on, and beam levels get moved constantly — a new SKU is taller, a customer changes carton size, someone needs one more level for the peak season.

Moving a beam level is not an operational decision to be taken by whoever is free that afternoon. It changes what the run can carry, which means it changes what the load notice on that frame should say, which means the notice is now describing a system that no longer exists. A file that does not record modifications quietly stops describing the warehouse, and nobody notices until the day it matters. The same reasoning applies to mixing pallet sizes in a selective run — the configuration is part of the rating, not separate from it.

What an inspection record should actually say

This is where otherwise diligent operators lose the value of the work they are already paying for. An inspection happens, somebody signs a sheet, the sheet says the racking was inspected, and it is filed. A year later nobody can tell from it whether anything was wrong.

A record that is worth keeping says four things: the date; who carried it out; what was found, located specifically enough to find again — run, bay and level, not "some damage in the back aisle"; and what was done about it, and when. That last part is the one that gets omitted, and it is the one that matters most, because a record of damage found with no record of it being resolved reads worse than no record at all.

The same specificity is what makes the record useful operationally rather than only defensively. A bay that appears three times in four years is telling you something about a corner, a driver route or a door position — and that is a layout conversation, not a repair one. Impact patterns only become visible when the locations are written down consistently.

Who asks, and when

The list is longer than most operators expect:

• A landlord, at lease renewal or when reinstatement is being discussed — often together with floor loading.
• A client auditing a third-party logistics provider before placing work. This one is increasingly routine, and "we do not have that" is a scored answer.
• An internal audit, safety or quality.
• Anyone investigating an incident — a collapse, a serious near miss, a load that came down. Here the questions are narrow and specific: what was it rated for, and when was it last inspected.
• Your next installer, when you want to extend, reconfigure or move. Without records, that job begins with a survey that would not otherwise have been necessary.

Every one of those asks after the fact. That is the entire argument for keeping the file: none of these are moments when the paperwork can be created retrospectively.

Rebuilding a file that does not exist

Plenty of warehouses have no records at all — the racking was inherited with the unit, or installed by a company that has gone, or put up in stages by three different suppliers. It is recoverable. It is just not recoverable by guessing.

Capacity cannot be read off the steel by eye, and this is the single most important thing to be clear about: racking with no documentation has an unknown capacity, which is a different and more awkward situation than a low one. A low rating can be worked with. An unknown one cannot be worked with at all, only guessed at, and the guessing is done by forklift drivers loading it.

The route back is physical. Measure what is installed. Identify the components where markings survive. Establish a rating that can be stated rather than assumed. Then fit load notices that match it, and start the file from there — dated, with the survey as its first entry. That is also the honest starting point for anyone taking over a warehouse with racking already in it.

What good looks like

It is not a large amount of paper. One folder, physical or otherwise, containing: the handover pack, the as-built layout, the load figures behind the notices, every inspection report in date order, and a running note of modifications with dates. Someone named as responsible for it. And a habit of adding to it on the day something changes, rather than at the end of a year nobody can now reconstruct.

That is perhaps an hour a year of actual effort. Against it: the survey you would otherwise pay for, the audit you would otherwise fail, and the questions you would otherwise be answering from memory.

If your file is thin

Most are, and there is no particular shame in it — racking is the one asset in a warehouse that works silently for years and asks nothing of anybody. The problem is that silence is not evidence.

Tell us what is installed and what you can still find, and we will tell you what is missing and what it takes to establish the rest. If we installed it, we very likely still hold the drawings.

WhatsApp us on 9107 2601 — photographs of the frames, the beam connectors and any load notices already fitted are the most useful first message. If you are planning changes rather than chasing paperwork, our installation and inspection services start from the same survey either way.