Why EAC Matters More Than Any Other Metric
CPI tells you where you have been. SPI tells you how fast you are going. But the question every program manager, customer, and executive actually needs answered is: how much will this program cost when it is done? That is the Estimate at Completion (EAC).
EAC is the single most consequential number in EVMS. It drives budget decisions, triggers contract modifications, determines whether a program survives its next review, and shapes the careers of everyone involved. An unrealistic EAC — whether too optimistic or too pessimistic — causes bad decisions. Getting it right requires both mathematical rigor and professional judgment.
The four standard methods each make different assumptions about future performance. Understanding those assumptions is the key to choosing the right method — and explaining your choice to stakeholders.
The Four Standard EAC Methods
| Method | Formula | Assumption | Best Used When |
|---|---|---|---|
| 1. CPI-Based | EAC = BAC ÷ CPI | Future cost efficiency will match past cost efficiency | Program is past 20% complete; no major changes ahead; CPI is stable |
| 2. CPI Remaining | EAC = AC + (BAC – EV) ÷ CPI | Remaining work will be performed at the current CPI | Same as Method 1; mathematically equivalent when using cumulative CPI |
| 3. CPI × SPI Composite | EAC = AC + (BAC – EV) ÷ (CPI × SPI) | Both cost and schedule inefficiency will persist | Schedule delays are driving cost overruns (overtime, extended overhead, resource conflicts) |
| 4. Bottom-Up ETC | EAC = AC + Bottom-Up ETC | Past performance is not predictive; fresh estimate required | Major scope change, new team, different technical approach, or early in program life |
Methods 1 and 2 are mathematically equivalent when using cumulative CPI. Method 2 simply makes the logic more explicit: start with what you have already spent (AC), then estimate the remaining cost by dividing remaining work (BAC – EV) by your efficiency (CPI). Most practitioners prefer Method 2 for its transparency.
💡 The Christle Rule and CPI Stability
Research by the DoD (often called the “Christle study”) demonstrated that cumulative CPI stabilizes after approximately 20% of the work is complete. After that point, final CPI is almost always within ±0.10 of the current value. This is why CPI-based EAC methods are most reliable after the 20% completion milestone. Before that point, consider a bottom-up ETC or weight the statistical EAC with greater uncertainty.
Worked Example: Four Methods, One Dataset
Given Data:
| Metric | Value |
|---|---|
| Budget at Completion (BAC) | $10,000,000 |
| Earned Value (EV) | $4,500,000 |
| Actual Cost (AC) | $5,400,000 |
| Planned Value (PV) | $5,000,000 |
| CPI (cumulative) | 0.833 (EV ÷ AC) |
| SPI (cumulative) | 0.900 (EV ÷ PV) |
| Bottom-Up ETC (from CAMs) | $5,800,000 |
Method 1 — CPI-Based:
EAC = BAC ÷ CPI = $10,000,000 ÷ 0.833 = $12,005,000
Method 2 — CPI Remaining:
EAC = AC + (BAC – EV) ÷ CPI = $5,400,000 + ($10,000,000 – $4,500,000) ÷ 0.833 = $5,400,000 + $6,603,000 = $12,003,000
(Difference from Method 1 is rounding only — they are mathematically equivalent.)
Method 3 — CPI × SPI Composite:
EAC = AC + (BAC – EV) ÷ (CPI × SPI) = $5,400,000 + $5,500,000 ÷ (0.833 × 0.900) = $5,400,000 + $5,500,000 ÷ 0.750 = $5,400,000 + $7,333,000 = $12,733,000
Method 4 — Bottom-Up ETC:
EAC = AC + ETC = $5,400,000 + $5,800,000 = $11,200,000
| Method | EAC | Overrun vs BAC | Overrun % |
|---|---|---|---|
| 1. CPI-Based | $12,005,000 | +$2,005,000 | +20.1% |
| 2. CPI Remaining | $12,003,000 | +$2,003,000 | +20.0% |
| 3. CPI × SPI Composite | $12,733,000 | +$2,733,000 | +27.3% |
| 4. Bottom-Up ETC | $11,200,000 | +$1,200,000 | +12.0% |
Analysis: The four methods produce EACs ranging from $11.2M to $12.7M — a spread of $1.5M. Method 4 (bottom-up) is the most optimistic, implying the team believes future efficiency will improve to CPI ≥ 0.95. Methods 1–2 assume current CPI persists. Method 3 adds schedule-driven cost growth. The PM must decide: is there credible evidence to support the bottom-up optimism, or should the statistical range govern?
Management EAC vs. Statistical EAC
Programs typically maintain two EACs in parallel:
| Type | Source | Strengths | Weaknesses |
|---|---|---|---|
| Statistical EAC | Calculated from CPI, SPI, or CPI × SPI | Objective, repeatable, grounded in historical data; resistant to optimism bias | Cannot account for known future changes: approved scope modifications, new team members, different approach |
| Management EAC | Bottom-up ETC from CAMs + AC | Incorporates expert judgment, known risks, corrective actions, and upcoming changes | Susceptible to optimism bias; often lower than statistical EAC without adequate justification |
The relationship between these two numbers is itself a diagnostic tool. When management EAC is lower than statistical EAC, the team is claiming they will outperform their historical trend. That claim requires evidence: specific corrective actions, staffing changes, or scope reductions that justify the improvement. When management EAC is higher than statistical EAC, the team has identified risks or scope growth that the indices do not yet reflect.
A persistent gap where management EAC ≤ statistical EAC with no supporting rationale is a red flag. It often indicates the team is “managing to the number” rather than providing an honest forecast.
TCPI: The Feasibility Check
The To-Complete Performance Index (TCPI) answers: what CPI must the team achieve on all remaining work to hit a given target? It is the most powerful reality-check tool in earned value.
TCPI = (BAC – EV) ÷ (Target – AC)
Where Target is either BAC (“can we still finish on budget?”) or EAC (“can we hit the current estimate?”).
TCPI to BAC: (BAC – EV) ÷ (BAC – AC) = ($10M – $4.5M) ÷ ($10M – $5.4M) = $5.5M ÷ $4.6M = 1.196
The team would need a CPI of 1.20 on all remaining work to finish at $10M. Current CPI is 0.833. This is not feasible — finishing on budget is effectively impossible.
TCPI to Management EAC ($11.2M): ($10M – $4.5M) ÷ ($11.2M – $5.4M) = $5.5M ÷ $5.8M = 0.948
The team needs a CPI of 0.95 to hit their bottom-up EAC. Current CPI is 0.833. This requires a 14% improvement in efficiency — ambitious but not impossible if corrective actions are credible.
As a rule of thumb: if TCPI ≥ 1.10, the target is almost certainly unachievable. If TCPI is between 1.00 and 1.10, it is theoretically possible but requires sustained improvement. If TCPI ≤ 1.00, the target is achievable at or below current performance levels.
⚠️ The Optimism Trap
When management EAC is lower than every statistical method, and TCPI to that EAC exceeds 1.05, the program is likely exhibiting optimism bias. Government reviewers will compare TCPI against CPI and ask: “What specific actions justify performing 15% better than your track record on every remaining dollar?” If the answer is vague, expect the customer to impose a higher EAC.
🎯 The Bottom Line
Four EAC methods exist because no single formula fits every situation. CPI-based methods (1 and 2) work best after the 20% completion point when trends are stable. The composite method (3) captures schedule-driven cost growth. Bottom-up ETC (4) is essential when conditions have changed. Always compare management EAC against statistical EAC and use TCPI to reality-check the result. If your TCPI exceeds 1.10, your target is a fantasy. Next: Integrated Baseline Review — validating the plan before execution begins.
Stop reading, start modeling
Model your process flow, run simulations, optimize staffing with TOC math, and test your knowledge with 107 interactive checks — all in one platform.