PMP Formulas Cheat Sheet: Every Formula, Explained Simply

Most people walk into PMP® prep more scared of the math than they need to be. There are about a dozen formulas worth knowing, nearly all of them are grade-school arithmetic, and once you see how they connect you can stop memorizing and start reasoning your way to the answer.

Here's the good news if you're testing in 2026. The updated exam that launched in July leans harder into scenarios and judgment, and it's lighter on pure number-crunching than older versions. You won't sit down to a page of calculations. But the concepts behind these formulas still show up, often dressed as a situation where you have to know what a CPI of 0.85 actually means for your project. So learn the ideas, not just the symbols.

This is your PMP formulas cheat sheet, with the reasoning included.

Learn these four shortcuts first

Before any single formula, get these four rules into your head. They decode most of what follows.

• A variance subtracts. An index divides. Cost Variance and Schedule Variance are subtraction. Cost Performance Index and Schedule Performance Index are division. That one distinction sorts half the earned value formulas.
• Cost uses AC. Schedule uses PV. Anything measuring cost compares your earned value to Actual Cost. Anything measuring schedule compares earned value to Planned Value.
• For the indexes, above 1 is good and below 1 is bad. For the variances, positive is good and negative is bad. A CPI of 1.1 means you're under budget. A CV of negative 5,000 means you're over.
• Earned Value comes first every time. CV = EV - AC. SV = EV - PV. CPI = EV / AC. SPI = EV / PV. All four start from EV, the value of the work you've actually finished. Remember that EV leads, and you can rebuild the four core formulas even if your memory blanks.

Nail those four and the rest is mostly filling in the blanks.

Earned value management, worked all the way through

Earned Value Management (EVM) is where the formulas cluster, so let's run one project from start to forecast.

You're managing an app build. The total approved budget, your Budget at Completion, is $200,000. By this point on the calendar you planned to be halfway done. You're actually 40 percent done, and the finance report says you've spent $90,000 so far.

Start with the three inputs everything else is built from.

• Planned Value (PV) = planned percent complete x BAC = 0.50 x 200,000 = $100,000. That's the budgeted cost of the work you meant to have finished by now.
• Earned Value (EV) = actual percent complete x BAC = 0.40 x 200,000 = $80,000. That's the budgeted cost of the work you actually finished. Notice it uses the budget, not what you spent.
• Actual Cost (AC) = $90,000. This one you just measure. No formula, it's the money that left the account.

Now the health checks.

• Cost Variance (CV) = EV - AC = 80,000 - 90,000 = negative $10,000. Negative, so you're over budget by ten grand.
• Schedule Variance (SV) = EV - PV = 80,000 - 100,000 = negative $20,000. Negative, so you're behind, sitting on twenty thousand dollars worth of work you planned to have done.
• Cost Performance Index (CPI) = EV / AC = 80,000 / 90,000 = 0.89. For every dollar you spend, you're getting about 89 cents of planned value back.
• Schedule Performance Index (SPI) = EV / PV = 80,000 / 100,000 = 0.80. You're moving at 80 percent of the pace you planned.

So this project is both over budget and behind. Now forecast where it lands.

Estimate at Completion (EAC) answers the question "what will the whole thing cost if we keep going?" There are three common versions, and the exam usually hints which one to use.

• If current cost performance holds: EAC = BAC / CPI = 200,000 / 0.89 = about $225,000. This is the default when a question says trends will continue.
• If the overrun was a one-time fluke: EAC = AC + (BAC - EV) = 90,000 + (200,000 - 80,000) = $210,000. You've eaten the cost so far, and you finish the remaining work at the planned rate.
• If both cost and schedule problems will keep dragging: EAC = AC + [(BAC - EV) / (CPI x SPI)] = 90,000 + [120,000 / 0.71] = about $258,750. The most conservative forecast.

Then two quick follow-ups.

• Estimate to Complete (ETC) = EAC - AC = 225,000 - 90,000 = $135,000. That's the money you still need from here to the finish, using the first EAC.
• Variance at Completion (VAC) = BAC - EAC = 200,000 - 225,000 = negative $25,000. Negative, so you're forecast to blow the budget by about twenty-five thousand.

One more that trips people up: the To-Complete Performance Index, or TCPI. It answers "how efficient do we have to get to still hit the original budget?" TCPI = (BAC - EV) / (BAC - AC) = 120,000 / 110,000 = about 1.09. You'd have to run at 109 percent efficiency the rest of the way, after averaging 89 percent so far. That gap is the polite way of saying the original budget is probably gone.

You never needed anything harder than multiply and divide. The skill being tested is reading the story and knowing which number answers the question.

Quick check

Your project reports an SPI of 1.05 and a CPI of 0.92. What's the most accurate line for your status report?

A) Ahead of schedule and under budget B) Behind schedule and over budget C) Ahead of schedule but over budget D) On schedule and on budget

The answer is C. SPI above 1 means ahead of schedule, and CPI below 1 means over budget. The trap is A, which people pick when they see one number above 1 and assume both are healthy. Read the two indexes separately every time. SPI is the schedule story, CPI is the cost story, and they don't have to agree.

Three-point estimating

When a task is uncertain, you estimate it three ways: Optimistic (O), Most Likely (M), and Pessimistic (P). Two formulas turn those into a single number.

Say a feature comes in at 6 days optimistic, 10 days most likely, and 20 days pessimistic.

• Triangular, the simple average = (O + M + P) / 3 = (6 + 10 + 20) / 3 = 12 days. Every scenario counts equally.
• PERT, also called Beta = (O + 4M + P) / 6 = (6 + 40 + 20) / 6 = 11 days. This one weights the most likely value four times, so it pulls the estimate toward reality and away from a scary worst case.

They land a day apart, and that difference is the point of the exam question. PERT gives 11 because it trusts the most likely value. The plain average gives 12 because that far-out 20-day tail drags it up. Know which formula weights the middle and you'll pick the right one under pressure.

You can also measure how shaky an estimate is.

• Activity standard deviation = (P - O) / 6 = (20 - 6) / 6 = about 2.33 days.

A bigger spread between optimistic and pessimistic means more uncertainty baked in. Roughly one standard deviation on either side of the PERT number gives a range of about 8.7 to 13.3 days, which is a more honest thing to tell a stakeholder than a single fake-precise figure.

Communication channels

The number of one-to-one communication lines on a team grows faster than people expect, which is exactly why it gets tested.

Communication channels = n(n - 1) / 2, where n is the total number of people.

A team of 6 has 6(5) / 2 = 15 channels. Add one person for 7 total, and you jump to 7(6) / 2 = 21. One extra body, six new lines of communication to keep straight.

The classic trap is forgetting to count yourself. If a question says "you and your team of five," that's six people, not five. Read for the total headcount, then plug it in.

Expected monetary value

EMV puts a dollar figure on risk, and it's the backbone of decision-tree questions.

Expected Monetary Value = probability x impact.

Take two risks. There's a 30 percent chance a vendor slips and it costs you $50,000. There's a 20 percent chance you finish early and earn a $40,000 bonus.

• Threat: 0.30 x (negative 50,000) = negative $15,000
• Opportunity: 0.20 x (positive 40,000) = positive $8,000

Convention matters here. Threats are negative, opportunities positive. Add them for a net EMV of negative $7,000. That number is what you'd fold into a contingency reserve or compare across branches of a decision tree. When a question hands you a probability and a cost in the same breath, EMV is almost always what it wants.

The full cheat sheet

Everything in one place. This is the part to screenshot.

Metric	Formula (plain text)	How to read it
Planned Value (PV)	planned % complete x BAC	budgeted cost of work you planned to finish by now
Earned Value (EV)	actual % complete x BAC	budgeted cost of work actually finished
Actual Cost (AC)	money actually spent	measured, no formula
Cost Variance (CV)	EV - AC	positive is under budget, negative is over
Schedule Variance (SV)	EV - PV	positive is ahead, negative is behind
Cost Performance Index (CPI)	EV / AC	above 1 under budget, below 1 over
Schedule Performance Index (SPI)	EV / PV	above 1 ahead, below 1 behind
EAC (cost trend continues)	BAC / CPI	forecast total if current cost efficiency holds
EAC (overrun was a one-off)	AC + (BAC - EV)	finish the rest at the planned rate
EAC (cost and schedule trends continue)	AC + [(BAC - EV) / (CPI x SPI)]	most conservative forecast
Estimate to Complete (ETC)	EAC - AC	money still needed from here
Variance at Completion (VAC)	BAC - EAC	positive is under budget, negative is over
To-Complete Performance Index (TCPI)	(BAC - EV) / (BAC - AC)	efficiency needed to still hit BAC
Triangular estimate	(O + M + P) / 3	simple three-point average
PERT / Beta estimate	(O + 4M + P) / 6	weighted average, favors most likely
Activity standard deviation	(P - O) / 6	how uncertain the estimate is
Communication channels	n(n - 1) / 2	one-to-one links among n people
Expected Monetary Value (EMV)	probability x impact	risk value, threats negative

What this means for the 2026 exam

Worth repeating, because it should shape how you study. The current exam is more scenario-based than the version your study-group friends may have taken a year ago. You're far more likely to see "the CPI is 0.85, what do you tell the sponsor?" than "calculate the CPI." That's good news. It means understanding beats memorizing. If you know that a CPI below 1 means over budget and an SPI below 1 means behind, you can answer a lot of questions without touching a calculator.

Still, do the reps. Plug real numbers into these until the interpretation is automatic, because under exam pressure you want the meaning to surface instantly. The fastest way there is practice questions with worked explanations, which you can find free at pmprep.app/practice. Work a set, get the ones you miss explained, then come back and do it again.

The formulas were never the hard part. Knowing what they're telling you about a real project is the whole game.

PMP, CAPM, PMI-ACP, and PMI are registered marks of the Project Management Institute, Inc. pmprep.app is an independent study resource and is not affiliated with or endorsed by PMI.