The Bottom-Up ETC Process
A bottom-up ETC is the gold standard. It requires the CAM to examine every remaining work package and planning package, estimate the resources needed to complete each one, and sum the results. It is time-consuming but produces the most defensible estimate because it is grounded in the actual remaining work.
Inventory Remaining Work
List every open work package and every planning package in the control account. For each, determine: what scope remains, what resources are needed, and what risks exist. Do not skip planning packages — they hold budget for future work that must be estimated.
Estimate Each Work Package
For each open work package, estimate the hours by labor category, material costs, subcontract costs, and other direct costs needed to complete the remaining scope. Base estimates on current performance, not the original plan. If you are running at 0.85 CPI, do not estimate the remaining work at 1.0 efficiency.
Price the Estimates
Apply current labor rates (not planned rates, unless they are valid). Include any known rate changes. Apply current overhead and burden rates. Sum to get the total ETC for each work package.
Aggregate to the Control Account
Sum all work package ETCs to get the control account ETC. Add ACWP to get the EAC. Compare to the Budget at Completion (BAC) to determine the Variance at Completion (VAC = BAC – EAC).
Cross-Check Against Statistical Methods
Compare the bottom-up EAC against statistical EACs. If they differ significantly, understand why. Either the bottom-up has missed something, or the statistical method is not applicable to the remaining work.
Statistical EAC Methods
Statistical methods use cumulative performance data to project the final cost. They are not substitutes for bottom-up estimates, but they serve as powerful sanity checks. If your bottom-up EAC is significantly more optimistic than the statistical EAC, you need to explain why you believe future performance will improve.
| Formula | Calculation | Assumption | When to Use |
|---|---|---|---|
| EAC = BAC / CPI | Divides total budget by cumulative cost efficiency | Future cost efficiency will match past efficiency | Default cross-check. Most reliable after 20% complete. |
| EAC = BAC / (CPI × SPI) | Divides total budget by product of cost and schedule efficiency | Schedule delays are driving additional cost and will continue | When behind schedule and schedule is driving costs (overtime, idle resources, ripple effects). |
| EAC = ACWP + (BAC – BCWP) | Actuals to date plus remaining budget at planned efficiency | Future work will be performed at planned efficiency (CPI = 1.0) | Only valid when past problems are truly non-recurring and a specific corrective action has eliminated the cause. |
| EAC = ACWP + ETCBU | Actuals to date plus bottom-up estimate for remaining work | None — based on assessment of remaining work | Always the primary method. Others are cross-checks. |
💡 The Optimism Test
If your bottom-up EAC is more favorable than BAC/CPI, you are claiming that future performance will be better than past performance. This requires a specific, credible explanation: a corrective action that has been implemented (not just planned), a non-recurring problem that has been resolved, or a change in scope/approach. Without such an explanation, the statistical EAC is more likely to be correct. Research consistently shows that bottom-up EACs that are more optimistic than BAC/CPI are wrong more often than they are right.
Risk-Adjusted ETC
A single-point ETC ignores uncertainty. Risk-adjusted ETC acknowledges that the future is uncertain and provides a range or confidence-weighted estimate that accounts for known risks and opportunities.
| Approach | Method | Best For |
|---|---|---|
| Three-Point Estimate | Estimate optimistic (O), most likely (ML), and pessimistic (P) for each major element. ETC = (O + 4×ML + P) / 6. | Individual work packages with significant uncertainty. |
| Risk Register Overlay | Start with the baseline ETC. Add expected value (probability × impact) for each identified risk. Subtract expected value of opportunities. | Control accounts with identified, quantified risks. |
| Monte Carlo Simulation | Model cost elements as probability distributions. Run thousands of trials to produce a probability distribution of total EAC. | Program-level or large CA aggregation. Produces P50, P80 confidence levels. |
At the CAM level, the three-point estimate and risk register overlay are most practical. Monte Carlo is typically performed at the program level using inputs from multiple CAMs.
Reconciling Bottom-Up and Statistical Methods
The power of EAC development comes from comparing methods. When bottom-up and statistical EACs agree, confidence is high. When they diverge, the divergence itself is information.
| Scenario | Likely Explanation | Action |
|---|---|---|
| BU EAC ≈ BAC/CPI | Bottom-up and statistical methods agree. High confidence in the forecast. | Report with confidence. Note the convergence in your narrative. |
| BU EAC < BAC/CPI | CAM is more optimistic than history suggests. Claims future improvement. | Identify the specific corrective action or change that justifies improved efficiency. If none exists, revise the BU EAC upward. |
| BU EAC > BAC/CPI | CAM sees remaining work as harder than historical average. Possible scope growth or risk materialization. | Validate the BU estimate. If the increase is real, it may signal a need for MR or a baseline change. Flag to PM early. |
| BU EAC > BAC/(CPI×SPI) | Situation is worse than even the pessimistic formula suggests. Significant problems ahead. | Escalate immediately. This control account likely needs management intervention, MR, or restructuring. |
Defending Your EAC
Every EAC will be challenged — by the PM, by program control, and by the customer. A defensible EAC is built on transparency, data, and logic. Here is how to prepare for the defense.
✅ Defensible EAC
- Work package-level detail that sums to the CA total
- Current rates applied, not outdated planning rates
- Remaining scope clearly tied to WBS dictionary and SOW
- Efficiency assumptions based on recent performance, not hope
- Risks identified and quantified, either included or noted
- Reconciled against at least one statistical method
- Corrective action impacts specifically quantified
❌ Indefensible EAC
- “Same as last month” with no reassessment
- Assumes CPI = 1.0 for remaining work without justification
- Ignores known risks because “they might not happen”
- Uses planned rates when actual rates are significantly different
- Cannot explain the difference from statistical EAC
- Remaining hours estimated at a summary level without WP detail
🎯 The Bottom Line
The ETC is the CAM’s forecast of the future — and it must be both honest and defensible. Build it from the bottom up by estimating every remaining work package. Cross-check against statistical methods and explain any divergence. Account for risk. Update monthly and comprehensively quarterly. Remember: an EAC that is too optimistic does not save money — it delays the recognition of problems until it is too late to act. Next: Change Control — how to manage baseline changes when the plan must evolve.
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.