Stay aware of the annexures: Why construction contracts fail long before work starts

Construction disputes usually present as arguments about delay, defective work or payment. Look behind many of them, though, and the real cause can often be traced back to something that was never properly resolved before the contract was signed. A specification didn’t match the drawings, or a programme assumed something the contract didn’t actually provide for. Standard forms such as JCT and NEC provide the legal framework, but it is the project-specific documents sitting behind them, whether described as contract data, schedules, annexures or appendices, depending on the form of contract, that determine what the parties have actually agreed to build, deliver and pay for.

Understanding how those documents come to conflict with each other in the first place is the first step to avoiding the dispute they can eventually cause. Too often, parties devote most of their negotiating time to amending the standard form, while assuming the project documents themselves are broadly consistent. In practice, the opposite is frequently true.

Where inconsistency first appears

Construction projects aren’t generally procured in a straight line. A specification is drafted early, before design has settled, and drawings are revised, sometimes several times, as the design team responds to cost or planning feedback. Contractors price the documents issued at tender, only for value engineering to change parts of the design after pricing has been completed. A consultant updates a specification without necessarily flagging every change to the commercial team, and revised pricing is issued against drawings that have since moved on again.

None of this happens through carelessness. It is simply the reality of modern procurement, where legal, commercial and technical documents often develop in parallel rather than in sequence, and by the time lawyers are instructed to finalise the contract, the result is a stack of documents generated at different points in the project’s evolution, not all of which say the same thing about the same building.

Employer’s requirements specify a Category A finish, while the specification, drafted slightly later, refers only to a high quality finish. Drawings issued for tender are an earlier revision, and the contractor’s tender return excludes certain finishes on cost grounds, citing lack of clarity. None of these four documents is, on its own, wrong. Taken together, they describe several different versions of the same building, and the executed contract does not clearly say which one was actually agreed.

When commercial assumptions drift

The contract particulars, sometimes called contract data, set out the commercial framework within which everything else operates, including the contract sum, the dates for possession and completion, retention percentages and the rates applied to variations. These figures need to match what was actually agreed at the end of negotiations rather than what was still on the table earlier in the tender process. A rate for variations agreed during negotiation but never updated in the contract particulars means the contract particulars are likely to take precedence over what was discussed during negotiations, since it is the contract particulars, not the negotiation history, that a court or adjudicator will look to.

That commercial framework is translated into the practical mechanics of delivery through the preliminaries, which cover site access arrangements, sequencing, and the assumptions the contractor priced its programme against. A recurring pattern involves programme and possession. The contract particulars state a single date for possession of the site. Priced at tender, the preliminaries assume phased access to different parts of the site to allow early trade packages to start work, yet the employer’s requirements, drafted before the phasing strategy was finalised, are silent on phased possession altogether. Nobody has done anything obviously wrong, but the contractor’s programme and the employer’s legal entitlement to give possession do not match, and that mismatch tends to surface at the worst possible moment, once the contractor is already mobilised and delay is starting to accrue.

When negotiations never make it into the contract

Tender and clarification correspondence often captures the qualifications, exclusions and agreed departures raised during negotiation, and this is where a great deal of genuine agreement between the parties can end up living outside the contract itself. Where those items are left unresolved rather than reconciled into the contract conditions, a contractor can find itself relying on a qualification the employer believed had been superseded, while the employer relies on conditions the contractor believed had been amended by that same qualification. Both readings can be entirely reasonable, which is exactly why the dispute becomes difficult to resolve without reference back to the negotiation itself.

Precedence clauses are often treated as the safety net here, and to an extent they are, but they solve the wrong problem. A precedence clause tells an adjudicator or a court which document governs when two documents conflict. It does nothing to prevent the conflict arising in the first place, and by the time anyone is relying on it, the parties are already in dispute, the answer it produces may not reflect what either party actually intended, and the cost of getting there has already been incurred.

This is also where risk allocation quietly comes undone. Parties often spend considerable time negotiating caps on liability, indemnities, limitation periods and insurance obligations, and those provisions matter, but they assume everyone has the same understanding of the works being priced and delivered. Where the underlying documents are inconsistent, even carefully negotiated risk allocation provisions may offer little protection, because the parties end up arguing about what the contract actually required in the first place, rather than who carries responsibility for it.

When outside documents become contractual obligations

Third party and passthrough documents, such as leases, licences, warranties and site information reports, carry a different kind of risk, since they import obligations from beyond the immediate relationship between employer and contractor. A site investigation report attached without express wording confirming whether it is provided for information only, or may be relied upon, is a common source of latent conditions disputes, because a contractor facing unexpected ground conditions has an obvious incentive to argue that the report formed part of the basis on which the works were priced.

Similarly, where a lease or licence is passed through to a contractor, the specific obligations the contractor is actually required to comply with need to be identified expressly, since a general pass-through clause referring loosely to “the obligations in the attached lease” can leave real uncertainty about which provisions bind the contractor and which were only ever relevant to the original landlord and tenant relationship. Standard form documents sit in this category too, novation deeds, warranty forms, practical completion certificates, deeds of release, and they carry real legal consequences even though they are often treated as administrative attachments. Negotiating leverage over these documents is highest before the contract is signed and falls away sharply once the works are underway, so they deserve the same scrutiny as everything else, not a lighter version of it reserved for later.

Standard forms still depend on consistent information

JCT and NEC, in their unamended form, are not usually where these problems originate. Both are well established, extensively tested forms, and the market’s familiarity with them is part of why they remain the default choice for most projects. The standard form assumes the project documents are accurate. If they are not, the standard form cannot fix that, and amending the conditions to allocate risk differently, while leaving the underlying project documents inconsistent, does not solve the problem. It just means the parties have negotiated hard over a framework that is still being fed contradictory information.

Modern procurement has made consistency harder to maintain

The documents themselves have also become harder to keep aligned. Building Information Modelling and Common Data Environments mean drawings are revised and reissued continuously throughout a project, often faster than the contractual documents can be updated to match. A drawing revision uploaded to a CDE without a corresponding update to the specification or the contract particulars can leave the contractual position lagging behind the technical position by several revisions before anyone notices. Design responsibility matrices, now common on most projects of any scale, add another layer, since disputes increasingly arise not from the specification itself but from a mismatch between what the specification requires and what the matrix says a particular consultant or contractor was actually engaged to design. Funders and future purchasers add collateral warranties and third party rights schedules into the mix, and those warranties need to be checked against the underlying appointments and contracts to confirm the obligations being warranted actually match what was agreed, rather than a standard form warranty reused from the last project without being tailored to this one.

The interaction between these documents has become significantly more complex over the past decade, which is exactly why contract review can no longer be left to lawyers checking drafting in isolation. Technical teams are often best placed to spot discrepancies between drawings, specifications, programmes and design information. Commercial teams understand how pricing assumptions have evolved during procurement, while project managers can identify where sequencing or possession assumptions no longer reflect the reality of the programme. Effective review means each of these perspectives looking at the same package of documents, rather than each professional checking only the part they were sent.

A different approach to contract review

The instinct, when reviewing a construction contract, is to work through the documents one at a time: check the conditions, then the particulars, then the specification, then the drawings. That approach catches drafting errors, but it seldom catches the kind of inconsistency described throughout this article, because those inconsistencies live in the gaps between documents rather than within any single one of them.

A more effective review reads the documents together, asking at each stage whether what one document assumes is actually confirmed by the others. Does the specification say the same thing as the drawings, and is the programme in the preliminaries consistent with the possession dates in the contract particulars? Has every clarification raised during tender actually been reconciled into the conditions, rather than left sitting alongside them? Answering those questions properly needs the lawyer, the quantity surveyor, the project manager and the design team looking at the same package from their own perspective, at the same time, before anyone signs.

Construction disputes are often presented as problems that emerge during delivery, but many are set in motion much earlier. Where project documents have evolved separately during procurement without being properly reconciled before execution, the contract can contain uncertainty from the outset. Investing time in reviewing those documents before signing is seldom about making the contract longer or more complex. It is about ensuring that everyone is working from the same understanding of the project before significant time, money and resources are committed, which is almost always easier, quicker and significantly less expensive than trying to reconstruct that shared understanding once the project is underway and positions have already become entrenched.