The claim is built before the dispute

The industry’s expensive category error

Tier 1 contractors and major developers routinely treat delay and quantum as dispute functions. Planning manages a programme. Operations build the work. Commercial teams administer change and report margin. Experts arrive when the relationship has hardened, sometimes years later, and are asked to turn all of that into one defensible account.

The sequence is familiar and structurally unsound. It assumes proof can be manufactured retrospectively out of information produced for unrelated purposes. A programme update shows dates but not why logic changed. A cost ledger shows overspend but not which event caused it. A diary says access was restricted but does not identify the crew, the quantity or the activity. A notice reserves rights without connecting the event to a critical path or a cost code. Each system holds part of the truth. None can prove the whole claim.

The chain the project has to be able to walk

Contractual obligation, then event or instruction, then timely notice, then operational effect, then programme effect, then resource and productivity effect, then actual cost, then mitigation, then remedy. Delay analysis and quantum analysis occupy different parts of that chain. They cannot safely be merged and they must be coordinated.

Delay analysis asks whether and when an event affected a contractual completion or milestone obligation. Quantum analysis asks what recoverable cost the event caused. The first does not prove the second. An extension of time may protect against liquidated damages while producing no compensation. Non-critical disruption may reduce productivity without moving completion. A variation may be valued without establishing a single day of prolongation.

Technical guidance The SCL Delay and Disruption Protocol makes the same separation while encouraging contemporaneous assessment as the work proceeds. RICS guidance defines ascertainment of loss and expense as calculating, as precisely as the contract requires, the additional cost directly caused by another party’s default. Neither replaces the contract or the governing law. Both point at disciplined causation rather than end-of-project arithmetic.

The project record is a commercial control system

Records are usually described as protection if a claim arises. That undersells them. A well-designed record system improves the project before any dispute exists.

A costed daily allocation shows where labour and plant actually went. Set against planned quantities and production rates, it exposes emerging inefficiency while there is still time to act. A programme narrative explains why criticality migrated and whether resequencing recovered the time. A change register links instruction to scope, notice, forecast cost, programme activity and decision owner. A weekly event review forces the team to decide whether it is looking at a client-risk event, a contractor performance problem, or an unresolved combination of the two.

This is not an argument for more administration. It is an argument for better data architecture. A five-minute record carrying a stable event identifier can remove hours of later reconciliation. A hundred-page monthly report with no causal structure adds almost nothing.

Bid and contract: define what normal means

The best evidence of disruption is a reliable account of the performance that was promised, priced and planned, and that account begins before mobilisation.

The bid team preserves tender programmes, method statements, resource curves, productivity assumptions, quotations, clarifications, exclusions, access dates and risk allowances. The commercial team maps every material time and cost mechanism: notice triggers, conditions precedent, extension grounds, compensation events, programme duties, record requirements, valuation rules, concurrency wording, float, acceleration, suspension, liquidated damages, caps, exclusions, dispute steps, governing law and forum.

Planning tests whether the contractual programme is logically and operationally buildable. Operations own the method and the sequence rather than merely approving a planner’s abstraction. Commercial managers understand where the productivity assumptions sit in the price. If the baseline promises an access sequence operations never intended to use, it is a weak benchmark. If rates conceal labour assumptions that cannot be traced, a later disruption model starts in uncertainty.

The output is a basis-of-performance dossier: what was to be built, by whom, in what sequence, with which resources, on what access and design assumptions, at what planned production. It is the project’s control condition, and almost nothing else in this article works without it.

Mobilisation: design the evidence before events occur

Mobilisation establishes a common work breakdown across programme, cost, quantities, document control and site reporting. Perfect alignment is rarely possible; a translation table is mandatory. Activities, cost codes and locations need stable identifiers, and event numbering has to be common to notices, schedule narratives, cost accounts and dashboards.

The team agrees minimum record fields, owners and deadlines. A daily record identifies date and shift, location, activity, labour and plant, installed quantity, instruction or constraint, time lost, mitigation, photographs and the relevant event. The planner specifies how actual starts, finishes, progress, logic changes and calendars will be captured. Commercial management defines cost segregation and the evidence needed for daywork, variations, prolongation and productivity analysis.

Technology selection follows the information design, never the other way round. A platform can make capture easier, add audit history and connect records to a contract system. It cannot decide what the project needs to prove. A bad data model digitises confusion.

Execution: run an event-control loop

Every material variance enters a loop within twenty-four hours, and the loop has seven steps.

1. Observe. Operations records the factual change at the workface.
2. Classify. Commercial management identifies the possible contract mechanisms and the notice clocks they start.
3. Map. Planning identifies affected activities, logic, and the current critical and near-critical paths.
4. Measure. Operations and commercial record resources, quantities, idle time, rework and any changed method.
5. Respond. The team mitigates, seeks instruction or clarification, and updates the forecast.
6. Communicate. The authorised person issues the notice and the continuing particulars.
7. Close or escalate. Agree the event, reserve it with quantified uncertainty, or prepare a structured submission.

That loop changes what a claim register is for. It stops being a graveyard of disputed values and becomes a portfolio of live project decisions.

Forecasting: report ranges, not aspirations

Executive reporting tends to collapse uncertainty into a single claim value, and that number is frequently a negotiating aspiration rather than a risk-adjusted forecast. A mature report separates four quantities: recorded cost incurred, forecast cost to complete the affected work, the contractually submitted amount, and an independently assessed recoverable range.

It separates time the same way: current forecast completion, asserted extension, assessed extension, and exposure to contractor or neutral delay. Each event carries a confidence assessment for entitlement, factual causation, criticality, cost proof and collectability, and that confidence is evidence-based rather than a coloured opinion.

For a board, the most useful number is usually none of those. It is the evidence gap: the share of claimed value that cannot yet be linked to a reliable source record. That single metric directs attention while the gap can still be closed.

Closeout: reconcile before the memories leave

At closeout the team should not start the claim. It should reconcile the event graph it has already been maintaining: confirm actual completion and milestone dates, freeze native programme files, resolve open logic and data anomalies, reconcile event costs to the ledger, obtain subcontractor records, close provisional assumptions, and interview key people while they are still reachable.

Judgment A global claim may be legally permissible, as Walter Lilly & Co Ltd v Mackay [2012] EWHC 1773 (TCC) shows in England and Wales, but it remains evidentially risky and that decision turns on its own contract and facts. Complexity is not a substitute for causal work.

Method follows question and evidence

A delay method is a way of testing a proposition, not a badge of validity. A prospective time impact analysis asks what an event was likely to do at the contemporary status date. A windows analysis examines changing criticality across periods. An as-planned versus as-built comparison communicates broad variance and can conceal critical-path migration. A collapsed as-built tests a counterfactual by removing events from an as-built network, and is only as reliable as the logic somebody inserted into it.

Technical guidance AACE Recommended Practice 29R-03 presents forensic schedule analysis as a family of techniques whose selection depends on contract, facts, law and records rather than a prescriptive hierarchy. ASCE/CI 67-17 treats critical path, float, chronology, concurrency, responsibility and changed schedules as related technical questions. Courts still decide causation on admissible evidence.

The lifecycle advantage is not that it guarantees a perfect model. It is that it removes reconstruction. Native updates, explained logic changes, quantities, look-aheads and contemporaneous narratives let an analyst test what actually controlled the work. The difference between a credible windows analysis and an expensive animation usually lies in the monthly discipline that preceded it.

Cost is not causation

The ledger records expenditure, not responsibility. Quantum analysis has to establish why the additional cost was incurred, which event caused it, whether the contract permits recovery, and whether another cause or an existing allowance already explains it.

Prolongation, disruption and change stay separate. Prolongation concerns time-related resources retained through a compensable period. Disruption concerns lost efficiency: more input for the same output because conditions changed. A variation concerns altered scope and may carry direct cost, time effect and productivity consequences at once. Merging all three into delay cost invites duplication, and duplication is the first thing a competent opponent looks for.

Judgment Lucas Earthmovers Pty Ltd v AngloGold Ashanti Australia Ltd [2019] FCA 1049 shows the consequence of skipping the work. The expert adopted summary quantities without independent measurement, and the underlying workings and source explanations were absent. The Federal Court found deficiencies in proof. Experience did not substitute for measurement and verification.

The governance that makes integration real

Integration is not achieved by copying more people into emails. It needs authority and cadence. The project director sponsors a weekly time, cost and change review chaired jointly by commercial and project controls. Operations supply fact and mitigation. Planning explains path and forecast. Commercial explains clause, notice and valuation. Cost control reconciles the ledger. Document control verifies provenance. Decisions and dissent are both recorded.

• One event identifier across every system.
• One controlled chronology.
• One approved baseline with a transparent update history.
• One event-to-cost reconciliation.
• One executive view that distinguishes fact, forecast, claim and assessment.

External delay or quantum specialists can test the system early, and should not take over project management. The practical truth behind the integration argument is simple: a planner cannot quantify labour loss without resource and output evidence, and a quantity surveyor cannot attribute cost without the programme and the operational mechanism.

The executive conclusion

Claims capability is not measured by the size of a claims team. It is measured by the quality of project decisions and the traceability of their consequences. A sophisticated contractor or developer treats quantum and delay as lifecycle controls: define the planned bargain, capture variance at source, link notice, time, resources and cost, assess alternative causes, and report uncertainty honestly. The resulting evidence can support a claim or defeat one, and more usefully it can prevent the dispute by making consequences visible while they can still be managed.

The board question is therefore not whether you have enough records to win. It is whether your project can explain today, with evidence, why time and cost are changing. If the answer is no, the commercial risk already exists, whether or not anybody has called it a claim.