FAQs for Variations & Change

Question 1

What belongs in a variation register?

Accepted Answer

Every change, with its instruction reference, its status, its current value and its notice position, kept current enough to answer at a glance what is agreed, what is pending and what is at risk. A register earns its keep only if it can answer three questions instantly: what is agreed, what is pending and what is at risk. That takes more than a description and a number per change. It needs the instruction reference, the document or confirmation that created the change, so it traces to an authority rather than a memory. It needs a status, instructed, priced, submitted, agreed, so anyone can see how far along it is without opening the file. It needs a value, the agreed figure where there is one and your honest current assessment marked as an assessment where there is not, because a change with no number is one nobody can claim. And it needs its notice position: whether any time or notice window attaches, and whether the notice has been served, because a variation worth money and out of time is worth much less. Add the date each thing happened, keep it in one document rather than scattered across emails, and reconcile it against the payment application every month so the register and the account never drift apart. Narrative does not belong: the register is a control document, not a diary, and the moment it fills with commentary nobody keeps it current. Kept lean and live, it turns change from an argument waiting to happen into an account both sides already recognise.

Question 2

How do you price a variation before the work is done?

Accepted Answer

You measure the instructed change against the contract's valuation rules, forecast quantities from the drawings, and price the parts you cannot rate yet as clearly provisional, with any time consequence flagged separately. Pricing a change before it is built is not guesswork, it is forecasting against known rules, and the earlier you do it the more control everyone keeps. Start from the instruction and pin down exactly what has changed, before and after, from the drawings, so you price the change rather than the whole area around it. Take the quantities from the design: a variation not yet built still has drawings, and quantities measured from them are good until the as-built proves otherwise. Value those quantities the way the contract requires: the same work at the bill or activity-schedule rate, similar work pro-rata, and only genuinely new work at a fair rate built from first principles, naming the basis for each so it cannot be quietly swapped later. Where a part genuinely cannot be rated in advance, price it as an honest forecast and label it provisional, to be confirmed on the records once the work is done, rather than pretending to a precision you do not have. Flag any time or disruption consequence as its own separate head, because it proves differently and gets lost if folded into a rate. Pricing before the work gives both sides a number to agree or challenge while the change can still be planned, resourced and sometimes avoided, instead of discovering the cost as a fait accompli. A forecast price slightly wrong, corrected on the records, beats a perfect price that arrived a year too late to change anything.

Question 3

Do daywork sheets need signatures to be worth keeping?

Accepted Answer

A signed sheet is worth far more, because a signature records agreement to the resources at the time, but an unsigned contemporaneous sheet still beats a reconstruction, so keep every sheet and chase the signatures. The honest answer is that signatures matter a great deal and their absence does not make a sheet worthless. A daywork sheet records the labour, plant and materials a change actually consumed, and its power comes from being contemporaneous: written on the day, before anyone had a reason to remember it differently. A signature from the other side's representative adds something specific: it records that they saw and agreed the resources at the time, closing off the argument that the hours or the plant were never incurred. That is why a signed sheet is close to unassailable, and why chasing the signature day by day while the work is fresh is worth the friction it causes. But an unsigned sheet is not nothing. A contemporaneous record, consistent with the site diary, the allocation sheets and the delivery tickets around it, is still real evidence of what the change took, and it beats a figure reconstructed from memory at the final account. The discipline is twofold: keep every sheet, signed or not, because the record has value in its own right, and pursue the signature as routine, because it turns a strong record into an agreed one. Never treat an unsigned sheet as not worth filing, because the alternative to a flawed record is almost always no record, and no record is how good daywork claims quietly disappear.

Question 4

What happens to unpriced variations at month end?

Accepted Answer

They still go into the application, at your honest current assessment marked as an assessment, because a variation left off until it is perfectly priced is a variation the other side gets to forget. The instinct to leave a variation off the application until it is fully priced feels tidy and is quietly expensive. A change instructed and started is work the job owes you for, and its final price still moving does not mean its value this month is zero. The right move at month end is to carry every live variation into the application at your honest current assessment, clearly marked as an assessment rather than an agreed figure, and to refine it in later cycles as the measure and the rates settle. That protects two things. It protects cashflow, because you are paid something for the work as it proceeds rather than financing all of it yourself until the price is nailed down. And it protects the record, because a variation that appears on every application from the month it happened is far harder to dispute than one that materialises, fully formed and unheralded, at the final account, where its sudden appearance invites the question of why it was never claimed before. The discipline: mark assessments honestly, do not inflate them to bank a high opening position, and reconcile them against the register each month so the two tell one story. But the cardinal rule is that silence concedes. An unpriced variation that is never claimed is not kept in reserve, it is forgotten, usually by the party holding the money.

Question 5

Should variations be raised from site or from the office?

Accepted Answer

Both, at different moments: the site spots the change and flags it the day it happens, and the office confirms it in the contract's form and prices it, and the handover between them is where change gets lost. The question is really about a handover, and change gets lost exactly where the handover is missing. The site is where a variation is first visible: the extra dig, the revised detail, the instruction given on the scaffold, the condition nobody expected once the wall came down. If the people on site are not trained and expected to flag those the day they happen, the change is invisible until it is too late to prove. So the site owns spotting and flagging, promptly, in whatever quick form suits, a photo, a note, a message to the office, the moment something departs from the drawings or the priced scope. The office owns what happens next: confirming the instruction in the form the contract requires, entering it on the register, pricing it against the contract's rules, and carrying it into the application. Neither half works alone. A site that flags change with no office discipline behind it generates notes that never become confirmed, registered, priced variations. An office that runs an immaculate process but hears about change late is administering a version of the job weeks behind reality. The fix is a simple, known route from site to office: everyone knows what a change looks like and who to tell, and a standing office routine turns each flag into a confirmed, registered, priced line. Get that handover right and change is captured while it is cheap; leave it to chance and it surfaces, expensively, at the final account.

FAQs for Variations, Quantity Surveying

Frequently asked questions

Registers and capturing change

What belongs in a variation register?

What happens to unpriced variations at month end?

Should variations be raised from site or from the office?

Pricing and dayworks

How do you price a variation before the work is done?

Do daywork sheets need signatures to be worth keeping?