Two documents, one building
The contractor’s proposals are the tenderer’s answer to your employer’s requirements. They are prepared under time pressure, by people who have read your document faster than you would like, and they carry the contractor’s own drawings, specifications and clarifications.
They will not match your requirements, and that is not misconduct. Some divergences are the contractor solving a problem you left open. Some are a cheaper answer to a loose clause. Some are simple oversight. The only way to know which is which is to compare them line by line.
On most schemes the proposals arrive as a substantial package: drawings, outline specifications, a schedule of clarifications and often a list of assumptions. The volume is part of the problem, because a document that takes two days to read properly tends to get one afternoon at the point in the programme where it lands.
What the comparison finds
The recurring findings are consistent across schemes. Specification substitutions offered as equivalents, where equivalence is asserted rather than demonstrated. Scope moved into provisional sums, which converts a fixed price into an allowance. Clarifications that quietly reverse a risk allocation set out in the requirements.
There are also omissions, where a requirement is simply not answered. Silence in the proposals against a clear requirement is usually recoverable. Silence in both documents is a variation waiting to be instructed.
Why it is left undone
The reconciliation is skipped for understandable reasons. It arrives at the busiest point in the programme, when funding is closing and everybody wants signature. It is unglamorous work with no visible output until something goes wrong. And it feels like distrust at exactly the moment the parties are trying to start well.
The cost of skipping it is deferred rather than avoided. Every divergence that is not resolved before signature becomes a conversation later, conducted with one party already on site and the commercial position already fixed.
How the record is used later
A completed reconciliation is a short document listing each divergence, the requirement it departs from, and the agreed treatment: accepted, corrected before signature, or priced. It takes days to produce.
Two years later it is the paper that decides arguments, because it establishes what both parties knew and agreed at the point of contract. Reconstructing that position afterwards from correspondence is possible, expensive and rarely conclusive.
What the reconciliation document contains
The output is deliberately plain: a table with one row for each divergence found. Each row records the requirement it departs from, what the proposals say instead, the commercial effect where one can be established, and the agreed treatment.
There are only three treatments. Accepted, meaning the divergence is an improvement or is commercially neutral and the employer is content. Corrected, meaning the proposals are amended before signature so the documents agree. Priced, meaning the divergence stands and the value of it has been established and reflected in the contract sum.
What is not permitted is a fourth category, noted. A divergence that has been noticed and left unresolved is worse than one nobody spotted, because it establishes that the employer knew and proceeded anyway.
The document is signed off by both parties and attached to the contract record. That is what gives it force later, and it is why the exercise has to finish before signature rather than trailing into the first month on site.
What this means for you
Build the reconciliation into your programme as a task with a duration, not as something to be done if there is time. It sits between tender return and signature, and it needs somebody who will read both documents adversarially.
That work sits inside the employer’s agent role, and the measurement behind it inside client side quantity surveyor. Where the divergence is discovered late and has already become a claim, our scenario on contract interpretation disputes describes what can still be done.
There is a timing point worth planning for. The reconciliation cannot start until the proposals are complete, and it has to finish before signature, which means it sits in a window that is frequently squeezed from both ends. Allowing for it in the programme, with a stated duration, is what stops it being the task that gets dropped when the funding date moves. The exercise also produces a second output nobody expects, which is a list of things the contractor understood better than the client did. Those are worth accepting quickly and on the record, because a divergence that improves the scheme at no cost is the cheapest thing in the document and it tends to get lost in a review looking only for problems.