Earned Value Metrics Cheat Sheet

The Three Base Measures

Every other EVM number is built from these three. Each one is a dollar figure measured as of the same status date -- mixing dates across them produces numbers that look valid but aren’t measuring anything real.

The Four Derived Metrics

Two variances (in dollars) and two performance indices (unitless ratios), each comparing two of the three base measures above.

Worked Example

A fictional water-treatment retrofit, valued as of a single status date partway through construction:

The plan called for $500,000 of work to be done by this date. Only $450,000 worth actually got done, and it cost $420,000 to do it. Running the four formulas against those three numbers:

Read together: SV is negative and SPI sits below 1.0, so in cost terms the project is behind schedule -- less work got earned than was planned by this date. But CV is positive and CPI sits above 1.0, so the work that did get done was accomplished for less than its budgeted value. This is a real and common combination: a project can be behind schedule and under budget at the same time, because SV/SPI and CV/CPI are independent measurements. Nothing about one constrains the other.

SPI Is Not CPLI

SPI and SV are cost-based schedule measures -- they compare dollars of work earned against dollars of work planned. They say nothing at all about network logic, activity dependencies, or which specific activities are driving the project’s finish date. That’s a fundamentally different question from the one a Critical Path Length Index (CPLI) or total-float check answers, which is built entirely from the CPM network -- logic, durations, and float -- with no reference to cost or budget at all.

Because SPI is a dollar total across every activity in scope, a large volume of cheap, easy, non-critical work finishing on time or early can pull the index up to a healthy-looking 1.0 or better, even while a handful of expensive activities on the true critical path are quietly losing float. SPI has no way to weight those activities any differently than the easy ones -- it can’t see the network at all. A CPM-based check looking at float directly on the driving path would catch that problem immediately; SPI, computed from cost totals alone, would not.

Treating “on schedule” (an SPI reading) and “the critical path is healthy” (a CPLI or float reading) as the same claim is a common but real mistake -- they’re computed from entirely different data and can disagree.

Frequently Asked Questions

Is “percent complete” the same thing as EV?

Not by itself. EV is percent complete multiplied by the budgeted value of that scope -- 40% complete on a $1,000,000 activity is $400,000 of EV, but 40% complete on a $100,000 activity is only $40,000. Informal “percent complete” talk often drops the budget weighting entirely, which is exactly the kind of substitution this cheat sheet exists to head off.

Is SPI the same thing as being on the critical path?

No. SPI is a ratio of dollars -- earned value against planned value, summed across all the work in scope. It has no concept of logic, dependencies, or which activities actually drive the finish date. A project can show a strong SPI while the specific chain of activities controlling completion is in real trouble, or a weak SPI driven entirely by cheap, non-critical work that's behind.

Can CPI and SPI move in opposite directions?

Yes, and it's common. A project can be behind schedule (SPI below 1.0) while still under budget for the work it has completed (CPI above 1.0), or the reverse -- ahead of schedule but burning through budget faster than planned to get there. The two ratios are independent; nothing about one constrains the other.

Do PV, EV, and AC have to be measured as of the same date?

Yes. All three are only meaningful when measured as of the same status or data date -- comparing planned value from one reporting period against actual cost from another produces numbers that look like valid EVM output but aren't measuring anything real.