Project cost variance (PCV)

Cost variance shows whether a completed project used more or less money than its approved cost baseline. Historical reporting attributed striking overruns to major initiatives: a Standish Group summary stated that 70% of projects exceeded budget and that 52% finished at almost 200% of their initial estimate. Concorde was reported to cost 12 times its schedule, the Channel Tunnel 80% more than budget, and Boston’s “Big Dig” 275%—or $11 billion—over budget. The Big Dig had initially been priced at $2.8 billion; a Boston Globe estimate placed its ultimate interest-inclusive cost at $22 billion, with repayment continuing until 2038. Treat these figures as historical illustrations, not universal contemporary benchmarks.

When to use it

• Answer the key performance question: “To what extent are our projects delivered on budget?”

• Include the KPI in the operational processes and supply-chain perspective.

• Define the baseline, scope, collection process, formula, reporting frequency and data ownership before calculating the measure.

• Compare results with approved tolerances, comparable projects and the organisation’s historical forecasting performance.

Origins

Comparing actual project expenditure with an authorised budget is a foundational project-control practice. The simple PCV in this article is an end-of-project or like-for-like budget variance. It should not be confused with earned value management’s cost variance, which compares the budgeted value of work performed with its actual cost and can be used during delivery.

What it is

Perspective: Operational processes and supply chain perspective.

Key performance question: To what extent are our projects delivered on budget?

Material overruns can damage cash flow, investment capacity and reputation, and may create contractual or legal exposure. One historical example involved a lawsuit alleging a $20 million overrun on an Oracle software implementation.

PCV compares scheduled project cost with actual project cost for the same defined scope and price basis. A result of zero means actual cost matched the baseline; a negative result indicates an overrun; a positive result indicates expenditure below plan.

The number requires interpretation. A favourable variance may reflect efficiency, but it can also result from undelivered scope, lower quality, delayed invoices or an inflated baseline. An adverse variance may reflect poor control, an unrealistic estimate or an authorised scope change. Report the causes and forecast implications alongside the amount.

How to use it

Measurement

Freeze an approved cost baseline and define which costs it includes. Track authorised changes separately, accrue committed costs consistently and compare like with like. For active work, supplement the simple variance with forecast cost at completion and progress-based measures.

Data collection method

Extract the scheduled cost from the controlled baseline and the actual cost from the finance or project-cost system. Reconcile purchase orders, invoices, labour, accruals, contingency use and approved changes before calculation.

Formula

PCV = SPC − APC

Where SPC is the scheduled project costs

And APC is the actual project costs

For a portfolio or department, individual variances may be added to show the total monetary effect. Straight averages can mislead because a small project receives the same weight as a large one; use weighted rates or show the distribution when comparing performance.

Frequency

Monthly review is common, with more frequent monitoring for short, volatile or strategically critical projects. A completed-project variance should be finalised only after material costs and accruals are captured.

Source of the data

Use the controlled project baseline, change log, project-management system and financial records. Where systems are not integrated, reconciliation and clear cut-off rules are essential.

Cost/effort in collecting the data

Calculation is inexpensive when the scope baseline, commitments and actual costs are maintained consistently. Manual collection becomes costly when coding differs between systems, changes are not controlled or invoices arrive late.

Target setting/benchmarks

A result close to zero suggests accurate planning and controlled delivery, but the appropriate tolerance depends on uncertainty, project type and stage. Negative values indicate overspend. Positive values should be investigated rather than automatically celebrated, because they may expose weak estimating or incomplete delivery.

Example

A division is running three projects:

PCV Project A = 500,000 − 600,000 = −100,000

PCV Project B = 250,000 − 270,000 = −20,000

PCV Project C = 75,000 − 70,000 = 5,000

PCV Department = (−100,000) + (−20,000) + (5000) = −115,000

The department is therefore over its combined baseline. Management should still examine each project separately because the favourable result in Project C does not remove the causes or consequences of the overruns elsewhere.

Top practical tip

A ratio can make differently sized projects easier to compare. Under the article’s convention, PCPI relates scheduled cost to actual cost: below 1 indicates over-budget performance and above 1 indicates spending below plan. Define the convention prominently because standard earned-value practice uses different names and inputs.

Top pitfall

Do not wait until completion. For long projects, calculate variance and forecast at meaningful milestones, preserve authorised baseline changes and interpret cost together with schedule, delivered scope and quality. A project can appear under budget simply because work is late or missing.

Further reading

http://management.energy.gov/documents/performance_measures_final.pdf

www.ajdesigner.com/phpearnedvalue/schedule_variance_equation.php