Cost Variance and Schedule Variance Formula Explained

CV = EV - AC. SV = EV - PV. Two subtractions that tell you whether a project is over budget or behind programme - in pounds, not percentages, not ratios. Simple enough. And yet most commercial teams either ignore these numbers or misinterpret them at every period end.

Before you can calculate either, you need three base values from the earned value formulas: Planned Value (PV), Earned Value (EV), and Actual Cost (AC). Get those inputs wrong and the variances are fiction. The hard part isn't the maths - it's getting the inputs right.

What Are Cost Variance and Schedule Variance?

Cost variance and schedule variance convert project performance into a single number. Positive means good, negative means trouble. No percentages, no ratios - just a raw figure in pounds that tells you how far off you are.

I've sat in monthly project review meetings where a commercial manager presents a cost variance of -£180,000 and the project director nods and moves on. No follow-up. No root cause. No action.

£180,000. On a £25M job running at 4% margin, that's 18% of the entire profit gone in one period. Variances are an early warning system. A negative CV at period 4 on a 24-month programme gives you 20 months to fix it. The same negative CV at period 20? That's a post-mortem.

Cost Variance (CV)

Earned value minus actual cost. The result tells you whether you've spent more or less than the budgeted value of work completed.

• CV positive: Under budget. The work you've done cost less than you planned.
• CV negative: Over budget. You've spent more than the budgeted value of completed work.
• CV zero: Exactly on budget. Enjoy it. It won't last.

Here's the bit that trips people up: CV doesn't tell you whether the project is over budget. It tells you whether the work completed so far cost more than it should have. If you've done 30% of the work and spent 40% of the budget, your CV is negative - but you haven't necessarily overspent against the project timeline. You've overspent against the work done.

Schedule Variance (SV)

Earned value minus planned value. This tells you whether you've completed more or less work than scheduled for this point in time.

• SV positive: Ahead of programme. More work done than planned.
• SV negative: Behind programme. Less work done than planned by this date.
• SV zero: Exactly on programme.

Here's what catches people out: schedule variance is measured in pounds, not days. An SV of -£200,000 doesn't tell you you're a certain number of weeks behind. It tells you the value of work you haven't completed versus what was planned. Converting that into actual weeks requires looking at the planned rate of spend - which is where S-curve tracking comes in.

Why express time in money? Because it lets you compare schedule and cost performance on the same scale. When CV is -£500,000 and SV is -£200,000, you can immediately see the bigger problem is cost, not programme.

Why Variances Matter More Than You Think

Most commercial teams calculate variances. The problem is nobody acts on them. Variances are your earliest warning - arguably the earliest you've got. Combined with CPI, SPI, and forecasting metrics like EAC, they tell you the trajectory.

A CV of -£50,000 alongside a CPI of 0.92 doesn't just mean you're £50K over. It means that for every £1 of budgeted work you complete, you're spending £1.09. That inefficiency is baked into the job unless something changes. That's the real power of variance analysis. Not the snapshot. The trend.

Worked Example: A52 Highway Widening Scheme

A52 Carriageway Widening. NEC4 Option C target cost. BAC £18,500,000. 18-month programme. Reporting at Period 7 of 18. Planned programme shows 38% of work should be complete. Physical progress assessed at 33%. £7,200,000 spent to date.

Step 1: Calculate the three base values

Step 2: Calculate variances

Two problems. The project's £1,095,000 over budget on the work completed, and £925,000 behind programme. CV tells you the efficiency problem. SV tells you the productivity problem. Both negative at the same time? That's a job in trouble.

Step 3: Calculate indices for context

A CPI of 0.848 on a highways job at Period 7 is serious. Without intervention, the Estimate at Completion using EAC = BAC / CPI would be £18,500,000 / 0.848 = £21,818,182 - a projected overrun of £3.3M. On an NEC4 Option C contract, the pain share mechanism means the contractor absorbs a significant chunk of that.

What should the commercial team do? At Period 7, there's still time. Break the CV down by cost code. Find out whether it's a systemic efficiency problem or a one-off hit. The variance tells you there's a fire. The cost code breakdown tells you which room it's in.

Period-on-Period Variance Tracking

A single-period variance is a photograph. What you really need is the film. Here's the A52 scheme across six periods.

Look at the trend. Period 2 was fine. Period 3 was the first crack - CV flipped negative while schedule was still marginally positive. That's your canary. By Period 4, both were negative and accelerating.

If this commercial team had acted at Period 3 - when CV first turned negative - they'd have had 15 months to correct course. Instead, by Period 7, they've lost half the programme with a deteriorating trend. Variance tracking isn't a reporting exercise. It's a trigger for action. For a full 12-month project walkthrough, see the construction EVM worked example.

CV and SV vs CPI and SPI: When to Use Each

Both. But to different people, for different reasons.

Absolute Variances (CV, SV) - Best For

Commercial teams and project boards. A commercial director doesn't want to hear "CPI is 0.85." They want to hear "we're £1.1M over budget." Variances speak the language of money. They connect directly to margin, cash flow, and the P&L. They're also ideal for threshold-based escalation: "Any CV exceeding -£500,000 or -5% of BAC escalates to the project director."

Ratio Indices (CPI, SPI) - Best For

Trend analysis and forecasting. CPI and SPI are dimensionless ratios - comparable across projects regardless of size. A CPI of 0.90 means the same thing on a £5M job and a £500M programme. That universality is powerful. The EAC, ETC, and TCPI formulas all use CPI and SPI. You can't derive a meaningful forecast from absolute variances alone.

Reporting Variances to Project Boards

I've reviewed hundreds of earned value reports. Most fall into one of two traps: a wall of numbers with no narrative, or a single RAG status with no supporting data. Here's what actually works.

Every period-end board report should fit on one page and answer four questions:

1. Where are we? - Current CV, SV, CPI, SPI
2. Where are we heading? - EAC and projected overrun or underrun
3. What changed this period? - Incremental variances and the reasons
4. What are we doing about it? - Corrective actions with owners and deadlines

For a complete 9-section reporting structure, see the earned value report template.

Traffic light thresholds: Set these at project setup, not midway through. Standard thresholds: green within 2% of BAC, amber at 2-5%, red beyond 5%. But the absolute number matters too. On a £200M programme, 2% is £4M - that might need immediate attention regardless of the percentage. Set both percentage and absolute thresholds.

Common Mistakes With Cost and Schedule Variance

1. Confusing schedule variance with actual time delay - SV is measured in pounds, not weeks. An SV of -£500,000 doesn't mean you're a certain number of weeks behind. Want actual time delay? Look at the S-curve gradient or use Time Schedule Variance: SV(t) = ES - AT.
2. Treating variances as cumulative only - A cumulative CV of -£300,000 could mean you lost £50K per month for six months, or you lost £300K in month 1 and have been on target since. The corrective action is completely different. Always calculate incremental (period-on-period) variances alongside cumulative ones.
3. Ignoring positive variances - A positive CV doesn't always mean good news. On a target cost contract, a large positive CV might mean you're under-measuring progress rather than genuinely under-spending. I've seen teams celebrate a CPI of 1.15, only to discover the progress measurement was off - when it caught up, the CPI crashed to 0.92.
4. Using variances without a solid baseline - Garbage in, garbage out. If your planned value curve doesn't reflect the actual programme sequence, every variance is misleading. On one rail package, the PV was based on a tender programme that bore no resemblance to the agreed construction sequence. Every month the team reported positive SV. When someone re-baselined, SV flipped to -£1.4M overnight. For more, see the common earned value mistakes guide.
5. Not linking variances to specific cost codes - A project-level CV of -£200,000 is useless for corrective action. Break it down: is it earthworks? Drainage? Surfacing? Prelims? Without that breakdown, you're reporting a problem, not diagnosing it.

Reference Tables

Variance and Index Quick Reference

Variance Combination Diagnosis

Governance Thresholds

Plug your own project numbers into the earned value calculator to calculate CV, SV, CPI, and SPI instantly.