A project can be perfectly on budget and badly behind schedule at the same time — CPI and SPI measure two independent things, and a healthy one says nothing about the other.
Earned Value Management (EVM) gives project managers a way to answer "are we doing well?" with actual numbers instead of gut feel, but only if the two core efficiency indices are read separately. The Cost Performance Index (CPI = Earned Value ÷ Actual Cost) answers purely a cost-efficiency question: for every dollar actually spent, how much planned value did we actually earn? The Schedule Performance Index (SPI = Earned Value ÷ Planned Value) answers purely a schedule-efficiency question: of the work we planned to have done by now, how much did we actually get done? A CPI of 1.0 says nothing about schedule; an SPI of 1.0 says nothing about cost. Reading either one in isolation, or worse, averaging them into a single vague "health score," hides exactly the kind of trouble a project manager most needs to see early.
Earned Value (EV) — the budgeted cost of work actually performed — is the anchor both formulas share, but each divides it by a different second number. CPI divides EV by Actual Cost (AC), isolating whether the money spent produced proportionate value — a pure cost-efficiency question. SPI divides EV by Planned Value (PV, the budgeted cost of work scheduled to be done by this point) isolating whether the amount of work completed matches the plan's timeline — a pure schedule-efficiency question. Because AC and PV are two completely independent inputs (one from the accounting system, one from the baseline schedule), CPI and SPI can land anywhere on their own axis regardless of what the other is doing. A project running four crews of overtime labor to hit a slipping deadline is the textbook case: SPI recovers toward 1.0 as the extra labor catches the schedule up, while CPI craters, because that overtime labor costs far more per unit of earned value than the original plan assumed.
The two can move in completely opposite directions. Consider a project where the team has fallen behind and, to save money, has simply slowed the pace of work rather than adding resources: less work is being done, so actual cost stays low relative to plan — CPI can look fine or even excellent — while earned value badly trails planned value because the underlying work genuinely isn't getting finished on time, driving SPI well below 1.0. This exact pattern (good CPI, poor SPI) is a classic early sign of a project quietly slipping its finish date while its cost report looks perfectly healthy — which is precisely why PMP and EVM training insist on reporting both indices separately, on the same status report, rather than a single blended project-health number that could hide either problem behind the other.
Explains the Cost Performance Index (CPI = EV ÷ AC) and Schedule Performance Index (SPI = EV ÷ PV) in Earned Value Management, using worked numeric examples to show how a project can be on budget while badly behind schedule, or ahead of schedule while badly over budget.
Both indices are computed from Earned Value, both are reported on the same EVM status report, and both are unitless ratios centered on 1.0 — making them easy to skim as "two versions of the same health score." They measure genuinely independent things because they divide EV by two unrelated denominators: Actual Cost (a real accounting figure, dependent only on what was actually spent) and Planned Value (a schedule baseline figure, dependent only on what was scheduled to be done by this date). A project's cost performance and schedule performance can diverge sharply and often do.
Planned Value (PV): the authorized budget assigned to scheduled work as of a given date — what should have been spent by now, per the baseline plan.
Earned Value (EV): the budgeted value of work actually completed as of that date — regardless of what it actually cost to complete.
Actual Cost (AC): what was actually spent to complete that work.
Cost Performance Index: CPI = EV ÷ AC. A CPI below 1.0 means the project is spending more than the value it is earning (over budget); above 1.0 means spending less (under budget).
Schedule Performance Index: SPI = EV ÷ PV. An SPI below 1.0 means less value has been earned than was planned by this date (behind schedule); above 1.0 means more (ahead of schedule).
EVM is required reporting on most US federal and defense engineering contracts (per ANSI/EIA-748 guidelines) and is core PMP exam content, precisely because a project sponsor needs both numbers, not a blend, to make the right intervention decision. A low CPI with a healthy SPI calls for cost-control action (tighter procurement, reducing scope, renegotiating a subcontractor rate) without necessarily touching the schedule. A low SPI with a healthy CPI calls for schedule-recovery action (adding resources, working overtime, re-sequencing tasks) — which, notably, will often worsen CPI as a direct trade-off, since schedule recovery techniques like overtime and crashing typically cost more per unit of value than the original plan.
Cost Variance (CV = EV − AC) and Schedule Variance (SV = EV − PV) are the absolute-dollar counterparts to CPI and SPI's ratios — CV/SV tell you the dollar amount of variance, while CPI/SPI tell you the efficiency rate. Both pairs are typically reported together: a small CV on a small project can represent the same underlying problem as a much larger CV on a bigger project, which is why the ratio (CPI) is often more useful for comparing performance across projects of different sizes.
Yes — Estimate at Completion (EAC) is commonly forecast as EAC = Budget at Completion ÷ CPI (assuming current cost efficiency continues), and a similar time-based extrapolation using SPI can forecast a likely finish date, though more rigorous schedule forecasting typically layers in critical path analysis as well rather than relying on SPI alone.
Not usually as a rigid target — small fluctuations around 1.0 are normal and expected on any real project. What matters more is the trend over time (a CPI steadily declining across several reporting periods is a much stronger signal than a single period's value) and whether the index stays within the tolerance band the project's management plan defines as acceptable.
No — EVM measures cost and schedule efficiency against the baseline budget and plan; it says nothing directly about the technical quality or correctness of the work being performed. A team could earn value efficiently while producing defective deliverables, which is why EVM is always paired with separate quality management processes, not used as a quality substitute.
Try our Engineering Management, Business & Professional Skills Studio
More calculators, simulators, and guides for this discipline.
Data Analysis Studio
✨ Premium ContentA structured, paid professional training program for this discipline.