How to structure a time impact assessment under NEC4
- May 25
- 20 min read
By Roman Bazelchuk | NEC Accredited Project Manager | APMG Project Planning and Control
Founder, NEC Planning Solutions Ltd
A compensation event arises on an NEC4 project. The contractor's planning team builds a time impact assessment. The assessment shows a four-week delay to planned completion. The quotation goes in. Two weeks later, the project manager rejects it. The reasons given are vague: "the assessment is not adequately supported," "the methodology is unclear," "the impact appears overstated." The contractor revises the assessment. It comes back rejected again, with different reasons this time. The cycle continues for two months. Eventually the project manager makes their own assessment under clause 64. The contractor's quotation showed four weeks. The project manager's assessment shows one week. That difference is now the contractor's problem to argue.
This pattern is one of the most common on NEC4 projects in the UK, and the reason is rarely that the contractor's analysis is wrong. The reason is that the assessment is not structured in a way that allows the project manager to verify it. The contractor knows what happened, knows what the impact is, and has built a programme model that supports the conclusion. But the assessment package that lands on the project manager's desk does not walk the project manager through the logic in a way that can be tested, challenged at specific points, and ultimately accepted.
A NEC4 time impact assessment is not a technical exercise that produces a number. It is a structured argument that demonstrates a number. The difference between assessments that get accepted and assessments that get rejected is rarely the underlying analysis. It is the structure of the package presented.
This article explains how to structure a time impact assessment under NEC4 so that it can be reviewed, challenged at specific points rather than rejected wholesale, and ultimately accepted. It covers what each element of the assessment must contain, how the elements connect, and where most assessments fall apart.
What a NEC4 time impact assessment is for
Under clause 62.2 of NEC4, a contractor's quotation for a compensation event must include details of the assessment for both cost and time. The time element shows the alterations to the accepted programme caused by the compensation event. Under clause 63.5, the delay to the completion date is assessed as the length of time that, due to the compensation event, planned completion is later than planned completion as shown on the accepted programme current at the dividing date.
These two clauses together define the time impact assessment. It is a forward-looking measurement that takes the accepted programme current at the dividing date as its starting point, applies the effect of the compensation event, and identifies how much later planned completion now occurs. The completion date moves by the same amount. The terminal float between planned completion and the completion date is preserved.
The assessment is prospective. It measures the forecast impact of the event from the dividing date onwards, not the retrospective actual impact after the event has played out. This is a deliberate design choice in NEC. The contract rewards contractors and project managers who assess events while they are still live, when the parties can act on the assessment, rather than waiting until after the works are complete and arguing about what happened from hindsight.
The prospective nature of the assessment has practical consequences. Time risk allowances and contingencies that existed in the accepted programme are preserved. Subsequent events that occur after the dividing date do not enter into the assessment of this event. The assessment is built using forecast logic, not actual logic, even if the contractor knows what subsequently happened. The article on NEC delay analysis and extension of time covers the prospective assessment principle in detail.
What this means structurally is that a time impact assessment is built from a known reference point (the accepted programme at the dividing date) using a known method (modify the schedule logic to reflect the compensation event), to produce a known output (the new planned completion date). Each of those three elements has to be visible in the assessment package.
The five elements of a properly structured NEC4 time impact assessment
A time impact assessment that the project manager can accept contains five elements. None is optional. Most rejected assessments fail because one or more is missing or weak.
The first element is the dividing date and its justification. The assessment must state, on the first page, the exact dividing date being used. Under clause 63.1 of NEC4, the dividing date is the date of the project manager's instruction or the contractor's notification of the compensation event, whichever produced the event. The dividing date is not a flexible date. It is a specific date defined by the contract, and the assessment must use that date.
The second element is the accepted programme being used. The assessment must identify which version of the accepted programme is being used, with the version number, the date of acceptance, and the data date. Under clause 63.5, this must be the accepted programme current at the dividing date. If multiple accepted programmes exist, the relevant one is the most recent that was accepted before the dividing date.
The third element is the compensation event fragnet. The assessment must show the specific changes to the programme caused by the compensation event. This is typically a fragnet: a small network of activities that represents the event itself, any new work the event introduces, any changes to existing activity durations, and the logic links from the event into the affected activities in the main programme.
The fourth element is the impacted programme. The assessment must show the accepted programme with the fragnet inserted, recalculated by the scheduling software to produce a new planned completion date. The difference between the original planned completion and the impacted planned completion is the time impact of the event.
The fifth element is the narrative. The assessment must include a written explanation of the cause and effect chain, the assumptions underpinning the assessment, the mitigation considered, and the residual delay after mitigation. The narrative is what allows the project manager to follow the logic and accept the assessment without reconstructing it themselves.
These five elements form a complete assessment package. An assessment that contains all five can be reviewed, challenged at specific points if the project manager disagrees with any of them, and ultimately accepted or partially accepted. An assessment missing any one of them gives the project manager grounds to reject the package wholesale rather than engage with its substance.
Step one: identify and justify the dividing date
The dividing date is the foundation of the entire assessment. If the dividing date is wrong, every subsequent step is wrong. If the dividing date is unclear, the project manager has grounds to reject the assessment without engaging with anything else in it.
Clause 63.1 of NEC4 defines the dividing date as the date of the project manager's instruction that triggered the compensation event, or the date of the contractor's notification under clause 61.3, depending on which initiated the event. Some events are initiated by project manager instructions (changes to the scope, instructions to stop work, decisions that change a key date). Other events are initiated by contractor notifications (encountering a physical condition, weather, employer's failure to provide access). The dividing date is whichever of these triggered the specific event being assessed.
In practice, the dividing date is sometimes obvious and sometimes not. A project manager instruction issued through CEMAR with a clear date and reference number produces an unambiguous dividing date. A contractor notification of a weather event referring to a continuous period of adverse weather may have a less obvious dividing date. The first day the threshold was exceeded? The last day? The day the contractor notified? The contract is precise: it is the date of the notification under clause 61.3. The contractor must have made that notification, and the date of that notification is the dividing date.
The assessment package should state the dividing date in a single sentence at the top of the document, followed by a brief justification: "The dividing date for this assessment is 14 March 2026, being the date of the contractor's notification under clause 61.3 (reference NCE-027) of the compensation event under clause 60.1(12) for unforeseen physical conditions encountered during piling works in the east substation."
That sentence is short. It does three things. It states the date precisely. It identifies the trigger event with a contract reference. It connects the dividing date to a specific clause 60.1 ground. The project manager reading this knows immediately what is being assessed and what reference point the assessment is using.
Step two: identify the correct accepted programme
The accepted programme used for the assessment must be the one current at the dividing date. This is straightforward in principle but produces persistent confusion in practice.
The correct programme is the most recent programme that was accepted by the project manager (or deemed accepted under NEC4) before the dividing date. If the dividing date is 14 March and the most recent accepted programme was Revision 6 accepted on 28 February, the assessment uses Revision 6. If a later revision (Revision 7) was submitted on 7 March but had not yet been accepted by 14 March, that later revision is irrelevant to this assessment. The contractual baseline is the last accepted programme.
This produces a counterintuitive result that contractors sometimes resist: a more recent, more accurate programme that better reflects the current state of the works may not be the one that should be used if it has not been accepted. The contract is explicit. NEC's own guidance (the FAQ on using the correct programme) confirms that a later programme issued after the dividing date is irrelevant to the assessment of an earlier compensation event, even if that later programme would produce a more accurate picture.
The practical implication is that contractors must keep the accepted programme current to maximise their position on compensation event assessments. If accepted programmes are routinely several months out of date, the assessment is always built against an outdated baseline, which weakens the contractor's position. The article on NEC clause 32 programme revision covers the revision discipline that maintains a current accepted programme.
The assessment package should state, immediately after the dividing date, the version of the accepted programme being used: "The accepted programme used for this assessment is Revision 6, accepted by the project manager on 28 February 2026, with a data date of 23 February 2026. Revision 7 was submitted on 7 March 2026 but was not accepted at the dividing date, and is therefore not used for this assessment."
This statement closes off a common ground for rejection. The project manager cannot challenge the choice of programme because the contractor has already shown they understood the rule and applied it correctly.
Step three: build the compensation event fragnet
The fragnet is where most assessments fail technically, even when they get the dividing date and accepted programme right.
A fragnet is a small network of activities that represents the compensation event and its effect on the programme. It contains several elements, each of which must be visible and labelled.
The first element is the compensation event itself. This appears as one or more activities representing what the event is. For a project manager instruction to add scope, this might be the additional work activities. For a physical conditions event, this might be the additional excavation, the redesign activities, and the procurement of additional materials. For an employer access failure, this might be a delay activity representing the period of denied access.
The second element is the logic links from the event into the existing programme. The fragnet must show how the event activities connect to the activities in the main accepted programme. These links determine how the delay flows through. Without explicit links, the schedule cannot calculate the impact.
The third element is any changes to existing activities caused by the event. If the event extends an existing activity duration (for example, a piling activity now takes 12 weeks instead of 8 because of the conditions encountered), the duration change must be visible. If the event re-sequences existing activities, the new sequence must be visible.
The fourth element is the constraint reflecting the dividing date. The fragnet should be anchored to the dividing date through an appropriate constraint, so that the schedule recalculation produces a result that reflects the event occurring at the right point in the programme.
A good fragnet is small. Five to twenty activities is typical. It is contained within a dedicated WBS node for the compensation event, identified by the CE number, with consistent activity coding. It is built so that it can be exported separately, reviewed in isolation, and inserted into or removed from the main programme without disturbing the surrounding logic. The article on Primavera P6 for NEC programmes covers the WBS and activity coding structures that support this.
A bad fragnet is large, sprawling, and intermingled with the main programme. It modifies activities throughout the schedule without isolating the changes. It includes work that is not actually caused by the event, or excludes work that is. It uses inconsistent activity codes that make it impossible to filter out the event-specific changes from the rest of the programme. When this happens, the project manager cannot review the event in isolation, which is grounds for rejection.
Step four: produce the impacted programme
With the fragnet built and inserted into the accepted programme, the schedule is recalculated. This produces the impacted programme: the accepted programme as it would have looked if the compensation event had been known when the programme was originally produced, with the event included.
The schedule must be recalculated using the same scheduling settings as the accepted programme. Calendar assignments, scheduling options (longest path versus total float for critical path identification), and progress handling must remain consistent. Changing these settings between the accepted programme and the impacted programme produces a different result that does not reflect the genuine effect of the event.
The impacted programme should be saved as a separate file, clearly labelled as the impact programme for the specific compensation event. It is not submitted as a replacement for the accepted programme. NEC guidance is clear that an impact programme produced for the purpose of a compensation event assessment is not itself a programme submitted for acceptance under clause 31 or 32. It is a working document for the assessment of one specific event.
The output of the impacted programme is a new planned completion date. The difference between the planned completion on the accepted programme (current at the dividing date) and the planned completion on the impacted programme is the time impact of the event. Under clause 63.5, this is the amount by which the completion date moves.
The assessment should report this calculation explicitly: "The accepted programme (Revision 6) shows planned completion on 18 December 2026. The impacted programme, with the compensation event fragnet incorporated, shows planned completion on 8 January 2027. The time impact of this compensation event is 21 calendar days. Under clause 63.5, the completion date moves by 21 calendar days."
That paragraph is the core output of the entire assessment. Everything else exists to support it.
Step five: write the narrative
The narrative is the element most contractors underweight, and it is the element that most often determines whether the assessment is accepted or rejected.
A time impact assessment without a narrative is a programme file with a number on the front. The project manager has to reconstruct the cause and effect chain themselves, work out which activities are affected and why, identify what assumptions were made, and verify that the conclusion is supported by the analysis. This takes time the project manager does not have. The result is often a rejection on procedural grounds, not because the assessment is wrong but because the project manager cannot follow it.
A time impact assessment with a strong narrative is a guided tour through the analysis. The project manager can read the narrative, understand the argument, identify the points at which they might disagree, and engage with those specific points rather than rejecting the whole package.
A good narrative covers five things.
First, it states what the compensation event is, with the contract clause that grounds it under clause 60.1, the date the event was notified, the dividing date for the assessment, and the accepted programme version being used.
Second, it describes the cause and effect chain. What does the event do? Which activities does it affect first? How does the effect flow through the programme to planned completion? This should be readable as prose, not a list of activity IDs.
Third, it states the mitigation considered. What re-sequencing was looked at? What adjustments to resource allocation were considered? What was implemented? What residual delay remains after mitigation? Under clause 63.8, the assessment assumes the contractor reacts competently and promptly. The mitigation statement protects the assessment against reduction on the basis that the contractor should have done more.
Fourth, it states the assumptions. Every prospective assessment requires assumptions about what would have happened. What is the assessment assuming about resource availability, weather, sequence of subcontractor mobilisation, productivity? Stating assumptions explicitly does not weaken the assessment. It strengthens it, because the project manager can engage with the assumptions specifically rather than with the conclusion in the abstract. The article on NEC programme acceleration and mitigation covers the mitigation discipline that feeds into this.
Fifth, it states the time impact and the resulting completion date movement, in the same form as the impacted programme calculation: 21 calendar days of impact, completion date moves from 18 December 2026 to 8 January 2027.
The narrative should be three to five pages of prose for a typical event. Long enough to cover the five elements properly. Short enough to read in fifteen minutes. Written in clear language, not technical jargon. The project manager should be able to read the narrative and form a view of whether the assessment is correct without having to open the programme files.
Common failures and how to avoid them
Five failures recur across rejected NEC4 time impact assessments. Each is avoidable.
Wrong dividing date. The contractor uses a different date from the one defined by the contract, often the date the contractor became aware of the event rather than the date of the formal notification, or the date the project manager actually instructed quotations rather than the date they should have. The fix is to apply clause 63.1 strictly: it is the date of the instruction or the notification, whichever initiated the event.
Wrong accepted programme. The contractor uses the most recent submitted programme rather than the most recent accepted programme. If the dividing date is 14 March and Revision 7 was submitted on 7 March but not accepted, Revision 6 (accepted 28 February) is the correct programme. Using Revision 7 is grounds for rejection.
Fragnet that is not isolatable. The contractor modifies activities throughout the programme without containing the event-specific changes in a dedicated fragnet. The project manager cannot extract the event-specific impact from the rest of the changes. The fix is to build every CE under a dedicated Compensation Events WBS node, with activity codes that mark CE-specific activities, so the changes can be filtered and reviewed in isolation.
Missing mitigation statement. The contractor presents the full impact without addressing what mitigation was considered. The project manager reduces the assessment under clause 63.8 on the basis that competent mitigation would have reduced the impact. The fix is to include a mitigation statement in the narrative, even if mitigation was not possible. State what was considered, what was implemented, and what residual delay remains.
Subsequent events conflated. The contractor includes the effects of events that occurred after the dividing date in the assessment of an earlier event. The assessment of CE-12 should reflect the position at the dividing date of CE-12, not the position now. Including subsequent events confuses the cause and effect chain and produces an inflated impact figure that is easy to challenge. The fix is to assess each event separately at its own dividing date, building each on the accepted programme current at that date.
These five failures account for the majority of rejected assessments. Avoiding them does not guarantee acceptance, but it removes the procedural grounds on which most rejections rest, leaving only substantive disagreement about the underlying analysis. Substantive disagreement is workable. Procedural rejection is not.
When subsequent events change the picture
A common question on long-running compensation events is how to handle subsequent events that occur between the dividing date and the implementation of the assessment.
The principle is straightforward. Each compensation event is assessed at its own dividing date, against the accepted programme current at that date. If CE-12 occurs on 14 March and CE-15 occurs on 22 March, CE-12 is assessed against the accepted programme current at 14 March, and CE-15 is assessed against the accepted programme current at 22 March. If CE-12 has been implemented before 22 March, the accepted programme used for CE-15 should reflect that implementation. If CE-12 has been notified but not implemented, the accepted programme used for CE-15 is still the most recent accepted programme, which does not include CE-12.
NEC4 clarified this in the 2019 update by explicitly requiring the assessment to take into account progress and other events between the previous accepted programme and the dividing date. This avoids the situation where the accepted programme is months old and the assessment is built against a baseline that no longer reflects reality.
In practice, the cleanest sequence is to assess events in chronological order of dividing date, implement each event before assessing the next where possible, and keep the accepted programme current through clause 32 revisions. The article on NEC4 multiple compensation events: best practice covers the sequential assessment approach in detail.
How the assessment connects to the cost element
A NEC4 time impact assessment is one half of the compensation event quotation. The other half is the cost assessment, which is built against the schedule of cost components and the dividing date principle from clause 63.1.
The two halves must be consistent. If the time impact assessment shows an additional six weeks of site activity caused by the event, the cost assessment should include the cost of those six weeks of preliminaries, supervision, and time-related plant. If the time impact assessment shows resource intensity changes during a specific period, the cost assessment should reflect those changes. Inconsistency between the two halves is one of the fastest ways to lose credibility with the project manager.
The discipline is to build the time impact first, then build the cost assessment using the time impact as input. The time impact tells you what the programme looks like with the event. The cost assessment tells you what that programme costs. Building cost first and then trying to support it with a time impact assessment usually produces inconsistencies that are easy to challenge.
For projects on Options C, D, and E (target cost and cost-reimbursable), the cost element uses the schedule of cost components. For Options A and B (priced contract with activity schedule or bill of quantities), it uses the short schedule of cost components unless the parties have agreed otherwise. The structure of the time impact assessment is the same in either case.
What good looks like in practice
A NEC4 time impact assessment that meets the standard above arrives as a single package on the project manager's desk. The package contains the narrative document (three to five pages), the accepted programme file (clearly labelled with version and acceptance date), the impacted programme file (clearly labelled as the impact programme for the specific CE), the fragnet extract showing the CE-specific activities and logic, and a summary calculation page showing the planned completion comparison and the resulting completion date movement.
The narrative leads with the headline conclusion, then walks through the five elements in order: dividing date, accepted programme, fragnet construction, impacted programme calculation, and mitigation statement. Each section is two to four paragraphs. Activity codes are referenced in the text where relevant but the prose is readable without needing to open the programme files.
The project manager reads the narrative in fifteen to twenty minutes, opens the programme files only to verify specific points, and either accepts the assessment or identifies specific points for discussion. The points for discussion are usually narrow: an assumption about resource availability, a logic link in the fragnet, the magnitude of mitigation considered. These narrow points can be resolved through dialogue rather than wholesale rejection.
This is what acceptance looks like. The contractor's analysis is sound, the structure of the assessment allows the project manager to verify it efficiently, and the narrow points of disagreement are resolvable. Most NEC4 time impact assessments fail to reach this standard not because the underlying analysis is weak but because the structure of the package does not allow the project manager to engage with it constructively.
For contractors without dedicated planning resource at this level, specialist NEC programme support provides time impact assessment as a managed service, building each assessment to a structure that meets the project manager's review needs and protects the contractor's commercial position.
Summary
A NEC4 time impact assessment is a structured argument, not a technical exercise. The difference between assessments that get accepted and assessments that get rejected is rarely the underlying analysis. It is the structure of the package presented.
A properly structured assessment has five elements: the dividing date with justification, the accepted programme with version and acceptance date, the compensation event fragnet built in isolation, the impacted programme with the time impact calculated, and the narrative that walks the project manager through the cause and effect chain.
Each element protects against a specific ground for rejection. The dividing date and accepted programme elements protect against procedural rejection on the basis that the wrong reference point was used. The fragnet protects against rejection on the basis that the event-specific impact cannot be extracted from the rest of the programme. The impacted programme protects against rejection on the basis that the time impact is not calculated from the schedule logic. The narrative protects against rejection on the basis that the project manager cannot follow the analysis.
Together, these five elements transform the assessment from a programme file with a number on the front into a structured argument the project manager can engage with constructively. Specific points can be challenged. Assumptions can be discussed. The assessment can be partially accepted, partially adjusted, or wholly accepted, but not procedurally rejected.
The discipline is not technically difficult. It requires understanding what each element of the assessment is for, building the assessment so each element is visible and verifiable, and writing a narrative that allows the project manager to follow the argument without reconstructing it from the programme files. Contractors who apply this discipline find that compensation event quotations move through acceptance faster, generate fewer cycles of rejection and revision, and produce stronger commercial positions when disputes arise. Contractors who do not find that even sound analysis fails the procedural test, and the project manager assumes control of the assessment under clause 64 with consequences that are usually unfavourable to the contractor.
About the author
Roman Bazelchuk is the Founder of NEC Planning Solutions Ltd, a UK project planning and controls consultancy supporting contractors with NEC programme compliance, compensation event assessments and live project controls. He is an NEC Accredited Project Manager and holds the APMG Project Planning and Control qualification, with a BSc in Mechanical Engineering and postgraduate training in Planning and Control.
NEC Planning Solutions provides contract-aware planning support through a QA-governed delivery model, helping project teams keep programmes accepted, current and commercially useful from tender through to live delivery.
FAQ
What is the dividing date for an NEC4 compensation event?
The dividing date is defined by clause 63.1 of NEC4 as the date of the project manager's instruction that triggered the compensation event, or the date of the contractor's notification under clause 61.3, depending on which initiated the event. The dividing date is fixed by the contract and not flexible. It separates work done before the dividing date (assessed using actual defined cost) from work not yet done at the dividing date (assessed using forecast defined cost).
Which accepted programme should be used for an NEC4 time impact assessment?
The accepted programme current at the dividing date. This is the most recent programme that was accepted by the project manager (or deemed accepted under NEC4) before the dividing date. A later programme submitted but not accepted is not the correct reference, even if it would produce a more accurate picture. NEC guidance is explicit on this point.
Is an NEC4 time impact assessment prospective or retrospective?
Prospective. Clause 63.5 measures the delay to the completion date as the length of time that planned completion is later than planned completion as shown on the accepted programme current at the dividing date, due to the compensation event. The assessment is built using forecast logic from the dividing date forward. Subsequent events that occur after the dividing date do not enter into the assessment of the event being considered.
What is a fragnet in an NEC4 compensation event assessment?
A fragnet is a small network of activities that represents the compensation event and its effect on the programme. It typically contains the event itself, any new work the event introduces, any changes to existing activity durations, and the logic links from the event into the affected activities in the main programme. A good fragnet is small, isolatable, and built under a dedicated WBS node so the event-specific changes can be reviewed separately from the rest of the programme.
What does clause 63.8 require for mitigation in a time impact assessment?
Clause 63.8 requires assessments to assume that the contractor reacts competently and promptly to the compensation event. In practice, this means the time impact assessment should consider what mitigation is possible through re-sequencing, reprioritisation, or resource reallocation within the existing programme, and present the residual delay after mitigation rather than the full unmitigated impact. A mitigation statement in the narrative protects the assessment against reduction by the project manager on the basis that the contractor should have mitigated more.
How does the time impact assessment connect to the cost element of the quotation?
The time impact assessment is built first and provides the input for the cost assessment. The cost assessment uses the schedule of cost components (or the short schedule of cost components on Options A and B) to assess the cost of the work caused by the compensation event, including time-related costs that flow from the time impact. The two assessments must be consistent: if the time impact assessment shows additional site activity, the cost assessment should include the cost of that activity.
What happens if subsequent compensation events occur between the dividing date and implementation?
Each compensation event is assessed at its own dividing date, against the accepted programme current at that date. NEC4 (in the 2019 update) clarified that the assessment should take into account progress and other events between the previous accepted programme and the dividing date. The cleanest practice is to assess events in chronological order of dividing date, implement each event before assessing the next where possible, and keep the accepted programme current through clause 32 revisions to maintain a reliable reference point.
Can the project manager make their own time impact assessment?
Yes, under clause 64.1, if the contractor's quotation is not accepted, or if the contractor fails to provide a quotation when instructed, the project manager can make their own assessment. The project manager's assessment uses their own assumptions about the contractor's planned timing, which almost always produce a lower entitlement than the contractor's quotation would have produced. The strongest defence against project manager assessment is a properly structured contractor quotation that the project manager can accept directly. The article on NEC clause 31 programme acceptance covers the broader procedural framework.
Need NEC4 time impact assessments structured to survive review?
If compensation event quotations are being rejected on procedural grounds, if the project manager is making their own assessments under clause 64, or if the time impact analysis is sound but the structure of the package is not protecting the contractor's position, specialist NEC programme support builds time impact assessments to the standard the project manager can verify and accept.
