Cost Variance vs Schedule Variance: What’s the Difference?

What is cost variance?

cost variance vs schedule variance is a key topic in modern project management and teamwork. Cost variance is the gap between the budgeted value of the work you have completed and the money you actually spent to complete it. It answers one question: for the work delivered so far, are we above or below the approved budget?

The formula:

*CV = EV − AC*

• EV (earned value) is the budgeted cost of the work actually finished by the status date. If a work package was budgeted at $40,000 and it is complete, the project has earned $40,000 regardless of the invoice.
• AC (actual cost) is the real money committed against that same work: labour, materials, subcontractors, licences, and allocated overhead.

A positive CV means the work cost less than budgeted. A negative CV means it cost more. Zero means the two matched, which is rare and occasionally a sign that nobody is recording costs properly.

Because CV is expressed in currency, it compares directly with a contingency reserve: a CV of −$24,000 against a $30,000 reserve is a warning, against $15,000 it is a breach. That is more actionable than “we are 11% over,” which is why absolute variance survives in reporting even where index-based metrics have taken over the analysis.

What is schedule variance?

Schedule variance is the gap between the value of work completed and the value of work the baseline said should be complete by that date. It answers a different question: are we ahead of the plan or behind it?

The formula:

*SV = EV − PV*

• PV (planned value) is the budget for all work scheduled to be finished by the status date. On a $720,000 project with a linear spend curve at month four of twelve, PV is $240,000.
• EV is the same earned value used in the cost formula.

Positive SV means more work is complete than planned. Negative SV means less. Zero means the project is tracking the baseline exactly.

Here is the part that confuses people in their first month of earned value: SV is measured in currency but describes time. It is the budgeted value of the schedule slip, not the slip itself. An SV of −$36,000 says the project earned $36,000 less value than the calendar required. Converting that into days takes another step that most status reports skip.

Cost variance vs schedule variance: what actually differs?

The difference is the second term in the subtraction. Cost variance compares earned value with actual cost. Schedule variance compares earned value with planned value. One judges spending, the other judges progress, and only one of them can be fixed with money.

The last row is the one worth memorising. CV keeps its meaning to the end: at completion it becomes the final underrun or overrun. SV is bounded by the plan. When the last task is finished, every dollar of planned value has been earned, so SV is zero. A project delivered eleven months late can show SV = $0 on the day it closes. Anyone using SV as evidence of schedule health in month eighteen of a twelve-month project is reading a metric that has already stopped measuring anything.

How do you calculate both in one reporting cycle?

You need a frozen baseline, a status date, a consistent progress rule, and costs booked against the same work packages. The arithmetic takes seconds; the data discipline takes the effort.

1. Fix the baseline. Use the approved cost baseline and note its version. If scope changes were approved, the baseline moves with them, on a recorded date.
2. Calculate PV for the status date. Sum the budgeted value of everything the schedule said should be done by that cut-off, not the total budget.
3. Calculate EV. Sum the budgeted value of work genuinely complete, using one rule per task type: 0/100 for short tasks, 50/50 for tasks under two weeks, measured quantities where a physical measure exists.
4. Collect AC for the same scope. Match costs to the period the work happened, not the invoice date. A prepaid annual licence for work spread over twelve months should not hit month one.
5. Subtract twice, then divide for percentages. CV% = CV ÷ EV and SV% = SV ÷ PV make work packages of different sizes comparable.

Worked example: a twelve-month platform migration

A company is migrating 60 internal services to new infrastructure. Approved budget is $720,000 over twelve months. At the end of month four:

Read it in plain language. The team has completed $204,000 of budgeted work and paid $228,000 for it, so it is $24,000 over budget on the work delivered. It has also completed 15% less work than the calendar demanded, which is $36,000 of value the plan expected and did not get.

Two facts, two different responses. The cost problem might be contractor rates, rework on two services, or a licence purchased early. The schedule problem might be an approval queue, a vendor API dependency, or three services harder than estimated. If the slip comes from a resourcing gap, adding people may fix SV and worsen CV. That trade-off is why you track both.

Why does the four-quadrant view matter more than either number?

Because the combination tells you what kind of project you are running, and the individual numbers do not. Four states are possible, and each has a characteristic cause and a characteristic trap.

Example: buying time you cannot afford

A marketing site rebuild has a budget of $150,000. At the end of week six, PV is $75,000, EV is $82,000, and AC is $97,000. CV = −$15,000. SV = +$7,000.

On a timeline slide this looks like a success: the work is ahead. On the ledger it is a warning. The team has spent $15,000 more than the earned value of the work delivered, and progress is only ahead because two contractors were added in week two. At the same efficiency, the final cost lands near $177,000 rather than $150,000, for a deadline that would probably have been met anyway.

That is the classic fast-and-expensive quadrant, and it is the most common way a project buys a schedule it did not need.

Are CV and SV enough on their own?

No. They are absolute amounts, so they do not travel well between projects of different sizes, and they do not forecast. Two companion ratios fix that.

CPI (cost performance index) = EV ÷ AC. Here, 204,000 ÷ 228,000 = 0.895: every dollar spent has bought 89.5 cents of budgeted value. CPI is the comparability layer, because 0.895 reads the same on a $720,000 project and a $72 million one. Our guide to the cost performance index covers interpretation bands, the estimate-at-completion forecast, and how the ratio misleads when costs are front-loaded.

SPI (schedule performance index) = EV ÷ PV. Here, 204,000 ÷ 240,000 = 0.85. SPI is useful for trend lines and cross-project reporting, and it shares SV’s weakness: it also converges to 1.00 at completion.

For a project in trouble, the pair also supports a forecast that uses both: EAC = AC + (BAC − EV) ÷ (CPI × SPI). In the migration that is 228,000 + 516,000 ÷ (0.895 × 0.85) ≈ $906,000 against a $720,000 budget, bleaker than the simple CPI forecast of $804,000 and closer to what happens when cost and schedule pressure compound.

How do you turn schedule variance into days?

Divide the variance by the planned value per period. The result, sometimes called time variance, converts a currency signal into the unit stakeholders care about.

For the migration, the planned value rate is $720,000 ÷ 12 = $60,000 per month. SV of −$36,000 ÷ $60,000 = −0.6 months, roughly 13 working days behind plan. A sponsor who hears “thirteen days” reacts differently from one who hears “−$36,000,” and the first sentence is the one that produces a decision.

The rigorous version is earned schedule. Instead of dividing by an average rate, find the point on the PV curve where planned value equals current earned value: here PV reaches $204,000 at month 3.4, and actual time is 4.0 months, so the slip is 0.6 months. Same answer, better method, and it matters most on projects with uneven spend curves where the average-rate shortcut drifts.

One caveat: earned schedule tells you how much time-equivalent value is missing, not whether the slip sits on the critical path. Ten days lost on a task with fifteen days of float costs nothing; three days lost on the critical path costs three days of the project. That check needs the network diagram, not the variance report.

Which problem should you fix first?

It depends on the contract type and on what the delay costs the business, not on which number is larger. That is the trade-off most guides avoid.

On a fixed-price contract, cost variance is largely the seller’s problem and schedule variance largely the buyer’s. The seller may accept a negative CV to protect a penalty date; the buyer may accept a slip to avoid a change order. Read both through the commercial terms before writing an action plan.

On time-and-materials work, every recovery has a price tag, because the client pays for the extra hours. A negative SV is the more urgent number there, since it converts into future cost.

On a project with a hard regulatory or revenue date, schedule dominates. A $40,000 overrun is trivial next to a fine or a quarter of deferred revenue. On a soft date with a firm budget, the reverse holds, and cutting scope is usually the only lever that fixes both variances at once.

The practical rule: fix the variance that drives the business outcome, and state explicitly which one you are accepting. A sentence such as “we are accepting a $20,000 cost variance to hold the launch date” turns two red numbers into a decision with an owner.

What are the most common mistakes?

• Using total budget as planned value. PV is the value that should have been earned by the status date, not the budget at completion.
• Allowing a positive SV off the critical path to reassure anyone. Easy non-critical work produces a healthy SV while the end date slips.
• Recording costs when invoices arrive instead of when work happens. This produces CV swings that reflect accounts payable, not performance.
• Recomputing the baseline to improve the picture. Rebaselining after approved scope change is legitimate. Rebaselining to make a report greener is not.

A related trap: both metrics inherit the quality of the earned value rule. A task claimed 90% complete that never reaches 100% inflates CV and SV by the same fiction.

Who should use these metrics?

Earned value fits projects with a defined scope, a stable baseline, and measurable work packages: construction, infrastructure, hardware, migrations, regulated programmes. It fits badly on discovery work and on operations with no finish line. The middle ground is a hybrid: run CV and SV at phase level rather than per sprint, base EV on accepted work rather than a “done” column, and keep contingency for discovery.

How do you keep both numbers trustworthy?

The recurring failure is not the maths. It is that the budget lives in finance, the schedule lives in one tool, and the actual hours live in another, so the monthly variance report becomes a manual reconciliation that arrives late and gets discounted.

Keeping planned value, earned value, and actual cost in the same workspace shortens that loop to minutes. That is the problem Doitify is built for: tasks, sub-tasks, owners, budgets, and progress reports in one place, so value earned and cost incurred are visible against the same work package. It is not an EVM engine for a capital programme with thousands of cost accounts, and does not try to be. For a 20- to 200-person team that needs a monthly CV and SV it can defend, having the inputs next to the work is the difference between a report and a ritual.

Conclusion

Cost variance and schedule variance are two subtractions that share one input and answer different questions. CV tells you whether the work delivered was worth what it cost. SV tells you whether enough work exists compared with the calendar. Neither tells you how late you are, and one of them stops telling you anything at all near the finish line.

Use them as a pair, convert SV into days before you talk to a sponsor, and read the combination rather than the individual numbers. Then write one sentence per reporting period explaining what moved and why. Six months later, that sentence will be worth more than either variance.

If this post on cost variance vs schedule variance was helpful, you might also enjoy All-in-One Project Management and CRM Software and Risk Register: What It Is and How to Create One.