CPI vs SPI is a comparison of cost efficiency and schedule efficiency within earned value management. The Cost Performance Index (CPI) shows how efficiently the program converts actual cost into earned value. The Schedule Performance Index (SPI) compares earned value with the budgeted value of work planned through the status date.
A value above 1.00 is favorable, while a value below 1.00 is unfavorable. However, the two indices answer different questions. CPI can support cost forecasting. SPI provides an early warning about the amount of planned work accomplished, but it does not measure calendar delay or replace Integrated Master Schedule (IMS) analysis.
CPI vs SPI at a Glance
Both indices use earned value in the numerator. Their denominators create the distinction:
- CPI = Earned Value ÷ Actual Cost
- SPI = Earned Value ÷ Planned Value
The legacy terms are Budgeted Cost for Work Performed (BCWP), Actual Cost of Work Performed (ACWP), and Budgeted Cost for Work Scheduled (BCWS). Therefore, the same formulas may appear as CPI = BCWP ÷ ACWP and SPI = BCWP ÷ BCWS.
The Defense Acquisition University EVM Gold Card provides these formulas along with other common earned value metrics. NASA’s program planning and control glossary also defines CPI and SPI as efficiency indicators.
What CPI tells the program manager
CPI measures the cost efficiency of completed work. For example, a cumulative CPI of 0.85 indicates that the program has earned $0.85 of budgeted value for every $1.00 of actual cost incurred.
- CPI greater than 1.00: The work accomplished cost less than its budgeted value.
- CPI equal to 1.00: The work accomplished cost exactly its budgeted value.
- CPI less than 1.00: The work accomplished cost more than its budgeted value.
CPI does not explain why the cost condition occurred. Labor inefficiency, material price growth, rework, overtime, accounting adjustments, learning-curve effects, or inaccurate planning can all affect the result. Therefore, managers should use the index to identify where deeper analysis is required.
What SPI tells the program manager
SPI measures the budget-weighted schedule efficiency of completed work. For example, an SPI of 0.85 indicates that the program earned $0.85 of value for every $1.00 of work it planned to accomplish through the status date.
- SPI greater than 1.00: More budgeted work was accomplished than planned.
- SPI equal to 1.00: Earned value equals planned value through the status date.
- SPI less than 1.00: Less budgeted work was accomplished than planned.
However, an SPI of 0.85 does not mean the program is 15 percent late in calendar time. SPI uses budget values rather than days, weeks, or milestone dates. It also does not account for activity criticality.
The Most Important Difference: SPI Is Not the Schedule
CPI directly compares the value of completed work with its actual cost. As a result, it provides a meaningful measure of historical cost efficiency. SPI compares two budgeted values, not baseline and forecast dates.
NASA’s Project Planning and Control Handbook states that schedule variance should serve as an indicator because it does not account for task criticality or measure time variance. The same limitation applies to SPI, which derives from the same earned value data.
For example, a program may earn substantial value by completing high-budget, noncritical hardware work early. That performance could produce an SPI above 1.00 even while a lower-budget software activity delays a critical integration milestone.
Therefore, program managers should compare SPI with:
- Critical and near-critical path status
- Baseline-to-forecast milestone movement
- Total float and negative float
- Late starts and finishes
- Missed baseline completions
- Schedule risk and margin consumption
- Driving predecessor logic
The IMS provides the time-based forecast that SPI cannot. For a broader explanation, see how to perform critical path analysis in an IMS and how to monitor near-critical paths.
How to Interpret CPI and SPI Together
Managers gain more insight by reviewing the indices as a pair. Each combination points toward a different line of inquiry.
CPI below 1.00 and SPI below 1.00
The program is earning less value than planned while spending more than the budgeted value of the work accomplished. This combination usually deserves immediate attention because both cost and budget-weighted schedule performance are unfavorable.
However, management should still identify the common root cause. A delayed engineering release, for example, may idle downstream resources, trigger overtime, and create both indices.
CPI below 1.00 and SPI above 1.00
The program is accomplishing work faster than planned in earned value terms, but at an unfavorable cost efficiency. Acceleration can produce this pattern when the team adds shifts, authorizes overtime, expedites material, or accepts lower productivity to protect a milestone.
Managers should determine whether the added cost represents a deliberate trade-off. They should also confirm that the early earned value comes from meaningful discrete work rather than an unsuitable measurement technique.
CPI above 1.00 and SPI below 1.00
The program has favorable cost efficiency but has earned less value than planned. Possible causes include understaffing, delayed material, deferred work, slower-than-planned hiring, or an optimistic baseline.
A favorable CPI may not remain favorable if unresolved delays later require overtime or schedule recovery. Therefore, the estimate at completion should reflect the likely cost of the forecast execution plan, not just historical efficiency.
CPI above 1.00 and SPI above 1.00
Both indices show favorable performance. Even so, managers should validate the result rather than assume the program has no execution risk.
Sustained favorable indices can be legitimate. However, they may also indicate front-loaded budgets, conservative planning, premature earning, accounting timing differences, or excessive Level of Effort (LOE). The DAU practical advice for reviewing EVM information warns that LOE can distort CPI and drive SPI toward 1.00.
Fictional Program Example
Consider a fictional avionics integration program with a Budget at Completion (BAC) of $40 million. At the current status date, the program reports:
- Planned Value: $10 million
- Earned Value: $8 million
- Actual Cost: $9.5 million
The calculations are:
- CPI = $8 million ÷ $9.5 million = 0.84
- SPI = $8 million ÷ $10 million = 0.80
The CPI indicates that the program earned about $0.84 of budgeted value for every dollar spent. Meanwhile, the SPI indicates that the program accomplished only $0.80 of work for every dollar of work planned through the status date.
Those numbers identify unfavorable conditions, but they do not identify the cause. The control account review shows that software integration started late because an interface design did not pass verification. The team then used overtime to recover test preparation, which reduced cost efficiency.
The IMS adds another critical fact: the integration event is six weeks late and drives the next system test milestone. The SPI alone did not express that six-week impact.
What the indices suggest about the forecast
If the program assumes cumulative cost efficiency will continue, a simple CPI-based independent estimate at completion would be approximately $47.5 million:
IEAC = BAC ÷ CPI = $40 million ÷ 0.84
A composite calculation using both CPI and SPI would produce a more pessimistic analytical result of about $57 million. However, neither formula should automatically replace the program’s bottom-up Estimate at Completion (EAC). Instead, the difference should prompt questions about remaining work, recovery costs, future risks, and the realism of management’s forecast.
For more on cost forecasting metrics, see TCPI, VAC, ETC, and advanced earned value management.
Current-Period and Cumulative Indices Answer Different Questions
A cumulative index describes efficiency from the start of the program through the current status date. In contrast, a current-period index isolates performance during the latest reporting period.
Cumulative values usually move slowly because prior performance remains in the calculation. As a result, a stable cumulative CPI can hide a recent deterioration. A poor current-period CPI may provide the first warning that rework, staffing changes, or supplier problems are affecting execution.
Current-period SPI can also reveal a missed monthly work plan. However, it may be volatile when the period contains a small amount of planned value or a few large work packages. Managers should therefore examine current, cumulative, and rolling trends rather than react to one isolated data point.
Why SPI Becomes Less Useful Late in the Program
Traditional cumulative SPI tends toward 1.00 as a program approaches completion. Once all planned scope has been completed, earned value and planned value converge, even when the program finishes late.
The Department of Energy’s project management lexicon distinguishes traditional SPI from a time-based Schedule Performance Index and notes this completion behavior. In practice, the limitation makes milestone forecasts, critical path analysis, and schedule risk information increasingly important during later execution.
Managers should not interpret a late-program SPI that improves toward 1.00 as proof of schedule recovery. First, confirm whether forecast completion dates have improved. Next, review remaining float, unresolved risks, and required work. Finally, determine whether the program has merely earned delayed work after its baseline dates passed.
Common CPI and SPI Interpretation Mistakes
- Converting SPI directly into calendar delay. An SPI of 0.90 does not establish that the program is 10 percent late or predict a specific number of delayed days.
- Using SPI instead of the IMS. SPI indicates budget-weighted accomplishment. The IMS identifies dates, logic, float, critical paths, and forecast milestone consequences.
- Reviewing only the program-level result. Favorable performance in one WBS element can offset poor performance in another. Drill down to control accounts and work packages.
- Ignoring Level of Effort. LOE earns value through the passage of time and can mask discrete schedule performance.
- Assuming a 1.00 index means no risk. A program can report neutral indices while facing supplier delays, future resource overloads, technical risk, or inadequate schedule margin.
- Treating every threshold as a regulation. A threshold such as 0.90 may support internal management or variance reporting, but no single trigger applies universally to every federal contract.
- Trusting the metric without testing the baseline. Poor logic, unrealistic budgets, weak objective measures, or inaccurate actual costs undermine the indices.
A credible Performance Measurement Baseline remains essential because CPI and SPI only measure performance against the plan represented in that baseline.
Are CPI and SPI Contractual Requirements?
CPI and SPI are standard EVM analytical metrics, but their use does not create a universal contractual threshold or corrective-action rule. Contract requirements depend on the applicable acquisition policy, clauses, Contract Data Requirements List, data-item descriptions, agency guidance, and program-specific tailoring.
FAR Subpart 34.2 addresses when an Earned Value Management System may be required and requires monthly EVMS reports for contracts to which EVMS applies. However, the contract and its reporting instructions determine the required content and level of detail.
Therefore, teams should separate three concepts: the contractual obligation to maintain or report EVMS information, the program’s approved variance-analysis thresholds, and management practices used to investigate trends. For an overview of the broader management system, see what an EVMS is and how it integrates scope, schedule, and cost.
Questions Program Managers Should Ask
- Which control accounts and work packages drove the index?
- Is the condition current, cumulative, or both?
- Did discrete work or Level of Effort generate the earned value?
- Does the IMS show an impact to a critical or near-critical path?
- What root cause connects the cost and schedule results?
- What is the effect on major milestones, the EAC, and management reserve?
- Does the corrective action address the root cause and appear in the forecast plan?
- Are the remaining budgets, durations, resources, and productivity assumptions realistic?
The strongest review does not stop at reporting that CPI or SPI crossed a threshold. It connects the metric to specific scope, identifies the execution problem, quantifies the downstream effect, and tests whether the corrective action can succeed.
The Bottom Line
CPI tells the program manager how efficiently the team has converted actual cost into earned value. SPI tells the manager how much budgeted work the team accomplished compared with the amount planned through the status date.
Use CPI to evaluate cost efficiency and challenge cost forecasts. Use SPI as a budget-based schedule warning. Then use the IMS to determine calendar impact, criticality, float, and forecast completion dates.
Neither index proves that a program is healthy or unhealthy by itself. Their value comes from trend analysis, valid underlying data, control account detail, and disciplined integration with technical, schedule, cost, and risk information.