Earned value can mislead on weak baselines

Earned value combines scope, schedule and cost through a baseline and a rule for recognising completed work. Its usefulness depends on that foundation. If the scope is incomplete, budgets are poorly distributed or progress rules reward ambiguous percentages, performance ratios can be precisely calculated yet misleading. A weak baseline does not become strong because its report contains an index. The first review should therefore test what work the reference represents and what evidence permits value to be recognised.

This chapter discusses the control logic qualitatively. The Delivery atlas has no cost baseline, planned value, earned value or actual cost inputs. It produces no earned-value indices or forecasts. The frozen schedule studies cannot be used to infer those quantities. A project that needs earned-value analysis should use an appropriate separately controlled model with scope mapping, time-phased budgets, progress rules and cost evidence. The conceptual distinctions here help a reader identify the prerequisites for that model without implying that the companion app performs it.

A schedule milestone and an earned-value event can coincide, but they need not. A supplier payment can occur before delivery acceptance. A design package can earn value under an agreed progress rule while a critical interface remains open. The performance system should explain these mappings. Otherwise favourable cost or value performance may be interpreted as readiness for a successor, which is a different claim. Integration requires a shared scope structure, not an assumption that every metric means complete in the same way.

A budgeted package should have a recognisable obligation and a progress rule suited to it. Discrete accepted deliverables can use evidence-based recognition. Ongoing support work may need a different treatment. The rule should be selected and documented before status pressure arises. Retrospective rule changes can make performance appear better without improving delivery. If a legitimate scope or method change requires a revised rule, preserve the earlier basis and explain the effect on comparability.

Unfinished work can dominate outcome even when much value has been recognised. A small remaining acceptance task may govern the service start. Conversely, a large budget package may be noncritical in the conditional schedule. A performance review should therefore show both economic progress and delivery release. Treating one as a substitute for the other conceals the causal structure. The sponsor needs to know what has been accomplished, what has been spent and which residual obligations still govern accepted completion.

An incomplete scope baseline can overstate progress by excluding work that later becomes essential. If commissioning preparation is missing, the remaining installation work may appear to represent nearly all completion while service readiness remains distant. Adding the missing obligation can make reported performance deteriorate even though the project has gained information. The correct response is to explain the baseline defect and its consequence. Suppressing the new obligation to preserve a favourable index weakens both the performance system and the delivery forecast.

Poor budget distribution can distort recognition. A package weighted heavily toward early procurement may show strong value progress before installation or acceptance, while another structure may recognise value later. Neither pattern is inherently wrong if it follows a justified scope and recognition basis. The danger is comparing ratios without understanding the mapping. A benchmark from another project is especially weak when scope, procurement strategy and recognition rules differ. The performance record should retain the basis needed for a fair comparison.

Actual costs can lag the work they describe. Accruals, invoices and payment timing may differ. A favourable cost report can therefore reflect incomplete cost capture rather than efficient delivery. The cost evidence should state its cutoff and treatment of incurred but unrecorded obligations. The schedule evidence needs its own cutoff. Comparing asynchronous records without reconciliation can produce an apparent performance trend that is partly a reporting artifact. This is a control issue requiring accounting and project evidence, not a correction the scheduling app can calculate.

A favourable schedule performance measure does not necessarily imply an earlier completion date. Recognised value may be distributed across noncritical work while a critical release remains delayed. The dependency network should be reviewed alongside the performance metric. The integrated manager can then explain whether progress is occurring where it changes accepted completion. This does not diminish the value of noncritical work. It prevents aggregate recognition from obscuring the remaining delivery mechanism.

Forecast-to-complete judgments should be based on remaining scope and credible execution plans. A mechanical extrapolation can be useful as a diagnostic comparison, but it should not automatically replace the package owner’s forecast when scope or method has changed. The record should show why the chosen forecast is suitable and what assumptions it depends on. When different approaches disagree, the disagreement is a prompt to examine the basis rather than an invitation to select whichever produces the most attractive result.

A fictional controller closes the cost report before the schedule team gathers its latest acceptance evidence. The executive pack compares both as if they share a status date. Some newly completed work has not yet entered the cost ledger, and some incurred obligations have not been invoiced. A favourable performance impression may partly reflect asynchronous evidence. The review identifies the mismatch without calculating a new index from incomplete data.

The controller states the cost cutoff and treatment of accruals. The schedule manager states the evidence cutoff and accepted-progress basis. The team reconciles material differences and identifies which comparison is suitable for the meeting. This does not require every record to be collected at the same instant. It requires a transparent basis and enough adjustment or qualification to prevent the audience from inferring a common period that does not exist.

The package map also differs. The cost report groups equipment procurement and logistics, while the schedule separates supply readiness and receiving acceptance. A single performance ratio cannot show which release obligation remains open. The controller and manager establish a mapping so economic progress and delivery readiness can be read together. The mapping does not force identical structures; it explains where their boundaries correspond and where they do not.

The sponsor asks whether the apparent favourable trend justifies a stronger delivery promise. The answer depends on residual work, acceptance and current resource evidence, not on the ratio alone. The team can acknowledge economic progress while retaining a conditional service forecast. This avoids both dismissing useful cost information and allowing it to stand in for a claim it cannot support. The report states what each lens establishes.

The review records a process improvement. Future packs will identify cutoffs, scope maps and material unrecorded obligations before drawing combined conclusions. This is a concrete control rather than a generic request for better reporting. Its effectiveness can be checked at the next close by inspecting the evidence fields and reconciliation. A later favourable metric still needs interpretation, but the basis will be easier to challenge.

The lesson is that performance arithmetic inherits the timing and scope of its inputs. A consistent formula cannot reconcile asynchronous evidence automatically. This case supplies no earned-value output or accounting opinion. It shows why the organisation needs a controlled bridge among progress, cost and forecast before treating their combined presentation as evidence of delivery health.

Imagine a fictional project reports strong budgeted-work progress because equipment production is well advanced. The operator still lacks an integrated test procedure and a readiness decision. The executive review should recognise the real production achievement while stating that the service outcome remains constrained by integration obligations. A single favourable headline would be incomplete. A single negative headline would also be misleading. The useful report connects progress to the remaining conditions for the accepted outcome.

The fictional controller completes a reconciliation note headed PRODUCTION-TO-SERVICE. The procurement account covers manufacture and transport; the schedule's supply event means equipment available for assembly; the operator's event means a tested service. The note records production recognition as supported, site compatibility as pending and service readiness as unresolved. It identifies an integrated test procedure that has no corresponding budgeted package. This is a missing obligation rather than a poor performance ratio. The repair is to locate its scope and funding in the control accounts before interpreting the aggregate trend. Merely moving value from procurement to commissioning would redistribute recognition without establishing the omitted work.

The package owners then review evidence rules. Production completion may support recognition under an agreed rule, but it should not mark receiving acceptance complete. Test preparation may earn progress before tests occur, while the test milestone remains open. These distinctions can be represented coherently if each rule is explicit. The executive report should use ordinary language to explain them. Technical abbreviations without the underlying obligation make it easier for readers to infer readiness that the metric does not establish.

The review then separates possible causes of a favourable cost reading. A missing supplier invoice directs the controller to incurred-cost capture. An early recognition rule directs the package owner to the earned-value basis. A delayed critical release directs the integration manager to the remaining network. These diagnoses can coexist; they should not be collapsed into a single explanation called efficiency. In the fictional disposition, production evidence is retained, the unrecorded logistics obligation is referred for accounting treatment, and the missing test procedure receives a scope assignment. The sponsor directs the procedure work because it governs service, rather than directing more already-advancing production to improve the headline.

PRODUCTION-TO-SERVICE closes with a contradiction test. If the cost ledger becomes complete while the service release remains blocked, the accounting repair has worked within its own boundary but has not removed the delivery constraint. If the test procedure is accepted while invoices remain incomplete, readiness has improved while the cost conclusion remains provisional. If both are repaired and the aggregate index barely changes, that does not negate either repair. Their value lies in different obligations. The note therefore carries separate statuses for economic recognition, incurred-cost capture and successor release. This produces an explanation of the disagreement between reports, rather than a demand that every report display the same colour.