Arguments about variation value are usually presented as arguments about rates and turn out to be arguments about facts: what was actually done, where, when, by how many people, under what conditions. Those facts live in site records or they do not exist at all, which is why records come before rates in variation claim preparation.
The records needed are not elaborate. They are the ordinary outputs of a well run site, kept in a form that somebody outside the project can follow.
Measurement of the varied work
The quantity, measured on the same basis as the contract, with the dimensions retained rather than only the total.
Measuring on a different basis from the bill guarantees an argument about quantity on top of the argument about rate, and it is entirely avoidable.
Where the varied work is covered up, measurement has to happen before it is, which means somebody has to be thinking about it at the time.
Allocation of labour and plant
Resources recorded against the varied work specifically, rather than against the project as a whole.
This single record converts most valuation arguments from assertion into evidence, and it is the one most often absent.
It matters most where daywork applies or where a star rate assumes particular outputs, because in both cases the assumption can then be checked against what happened.
A weekly summary is usually enough. Daily detail is better where the variation is substantial or contentious, and it costs very little more to produce while the work is happening.
The instruction itself
What was instructed, by whom, on what date, and under which provision. Confirmed in writing at the time rather than described later.
Where the instruction was oral, the confirmation that regularised it, sent within the period the contract allows.
Without this, the valuation question is premature, because the entitlement question has not been answered.
Conditions at the time
What access was available, what else was happening in the area, where the work sat in the sequence, and what the weather was doing if that is relevant.
These are the facts that support an adjusted rate, and they are the ones least likely to be recorded because they are obvious to everybody present.
A line in a site diary is usually enough. The point is that it exists, that it is dated, and that it was written before anybody was arguing.
Photographs
Dated, located and taken as a matter of routine rather than when somebody anticipates a dispute.
They are particularly valuable for work that gets covered up and for showing the state of an area before and after a change in sequence.
A photograph with no date and no location is much weaker than one with both, and most site photography now carries that data automatically.
What to do when records are thin
Establish what does exist before concluding that nothing does. Records are usually scattered rather than absent, and the subcontractor’s returns frequently show what the main contractor’s do not.
Look at documents created by others: inspection records, minutes, correspondence, the contract administrator’s own notes.
And be realistic. A variation valued at what can be shown recovers more than a larger figure that collapses under examination.
Keeping them usable
Per variation rather than per project. A folder for each instruction containing the instruction, the measurement, the allocation and the correspondence.
That structure takes minutes to maintain and it is the difference between an account that can be presented item by item and one that has to be reconstructed.
It also means a disputed item can be dealt with on its own rather than dragging the whole account into the argument.
What the other side keeps
Employers and contract administrators keep their own records, and those are evidence too. Inspection records, site visit notes and instructions issued all form part of the picture.
Where a contractor’s records are weak, the other side’s may fill gaps, and asking for them is reasonable rather than aggressive.
It also works in reverse. An assessor refusing a claim for lack of evidence should check whether their own records support or contradict it, because they will be produced eventually either way.
The record that proves the negative
Some variation arguments turn on showing that something was not in the original scope, which is harder than showing that it was.
The tender documents as issued, kept exactly as issued rather than as later revised, are what answer that question. Revised drawings overwritten in a shared folder destroy the evidence.
Keeping the tender set separately, untouched, costs nothing and settles a category of argument that otherwise runs for months.
Records the week after
Where a variation was carried out without proper records, the next best thing is a note made within days rather than months, by somebody who was there, describing what happened.
It is weaker than a contemporaneous record and far stronger than a reconstruction prepared for a claim eighteen months later.
The difference is that it was written before anybody knew it would matter, which is the quality that gives a record its weight.
Where the discipline comes from
Somebody deciding that it matters, before anything goes wrong. Record keeping improves sharply when the commercial team explains what the records are for.
It costs very little during the works and it is the difference between a variation account that settles and one that is argued about for a year.
That is commercial management work rather than claims preparation. Where an account has already stopped moving, the claims and dispute services start from whatever records exist rather than from the ones anybody wishes existed.
Variations that have not been agreed?
Variation claim preparation sets out the instructed scope, the measured difference and the valuation built on the rule the contract actually provides, with the evidence attached to each line.