Green Cost, Red Schedule
At the end of month six of a twelve-month programme the control account reports: planned value USD 2,400,000, earned value USD 1,950,000, actual cost USD 1,740,000, budget at completion USD 4,000,000. The sponsor sees the cost performance index of 1.12 and concludes the project is 'comfortably under budget', proposing to add a reporting dashboard to the release. The quality lead reports that rework hours have risen for four consecutive months and that 18% of completed work packages were signed off before all acceptance criteria were verified. The schedule performance index is 0.81 and three key milestones have slipped. The team is not idle; it is working on activities that are not yet complete but which have been partially credited in earned value.
1. Interpret the performance data. Is the project healthy, and which estimate at completion should you trust?
Read the worked answer
Run the numbers, then interrogate them. CV = 1,950 - 1,740 = +210 (under budget). SV = 1,950 - 2,400 = -450 (behind). CPI = 1.12. SPI = 0.81. Earned schedule is approximately 6 x 0.81 = 4.9 months, so about five weeks behind plan.
Three estimates at completion, three stories:
- Using CPI alone: 4,000 / 1.12 = USD 3,569,000. This is the number the sponsor is implicitly trusting.
- Using CPI and SPI together: 1,740 + 2,050 / (1.12 x 0.81) = 1,740 + 2,251 = USD 3,991,000.
- To-complete performance index at BAC: (4,000 - 1,950) / (4,000 - 1,740) = 2,050 / 2,260 = 0.91.
The judgement, not the formula. A cost under-run combined with a large schedule over-run, rising rework and premature sign-offs is the classic signature of earned value credited too early. If 18% of completed packages were signed off before acceptance criteria were verified, EV is overstated and the true AC of finishing that work has not yet been incurred. The 1.12 CPI may be measuring work not yet done and defects not yet paid for. When CPI and SPI diverge this sharply, the composite or a bottom-up re-estimate is the safer basis.
Conclusion: the project is not comfortably under budget. It is roughly five weeks behind, carrying an unpriced rework liability, and the honest planning figure is close to BAC - with the composite estimate showing essentially zero headroom. The 0.91 TCPI looks achievable on paper, but only if the underlying data is credible.
2. What corrective actions would you take in the next two weeks, and in what order?
Read the worked answer
First, fix the measurement before fixing the plan. Every recovery decision made on inflated EV will be wrong.
- Audit the 18%. Sample the prematurely signed-off packages and determine how much earned value was credited against work not actually accepted. Recomputed EV, CPI and SPI on the audited sample. This is a three-day exercise and it is the highest-value thing you can do.
- Change the earning rule. Move from partial-progress crediting to a defined rule - 0/100 for short tasks, fixed milestones for long ones - and require the acceptance criteria checklist before a package can be earned. This will make the next report look worse, which is the point.
- Price the rework. Four months of rising rework hours is a trend, not noise. Extract the hours and convert to cost and schedule; that number belongs in the EAC, not in a footnote.
- Recover the three slipped milestones selectively. Do not apply overtime across the board at CPI 1.12 - that is how you convert a cost under-run into an over-run. Identify which slipped milestone actually threatens downstream critical path, and compress only there, with the cost of compression compared against the cost of the slip.
- Re-baseline once, after the data is trustworthy, with the composite EAC and the audited schedule as the new reference.
What not to do now: start the dashboard, re-plan around the current EV, or announce a recovery date before the audit is complete. A recovery plan built on optimistic data fails twice - once on delivery and once on credibility.
3. How do you respond to the sponsor's proposal to add the reporting dashboard, given their reading of the CPI?
Read the worked answer
Separate the two claims and answer each on its own evidence. The sponsor has fused 'CPI is 1.12' with 'we have spare capacity'. Both halves need addressing, but not in the same sentence.
On the cost claim. Present the three EACs side by side - 3.57M on CPI, 3.99M on the composite, and the TCPI of 0.91 needed to finish at BAC even before rework is priced. The message is not that the sponsor is wrong; it is that the 3.57M figure depends on treating 18% of unverified sign-offs as complete work, and that assumption is under audit. Offer to revisit once the audit lands. Bringing the quality lead to that conversation, with the rework trend, is more persuasive than a debate about formulas.
On the dashboard. Treat it exactly like any other change: size it, log it, and price it against the composite EAC. At best the project is on budget with a rework liability and five weeks behind; there is no headroom to absorb new scope without a trade. The three honest options are: fund it from new money, descope an equal amount of baseline scope, or defer it to a later release.
On the relationship. Sponsors respond to a clear statement of what is known, what is uncertain and what decision is needed. Say: 'The cost position looks favourable, but it depends on data we are auditing, and the schedule is five weeks behind. I will bring you a verified position in two weeks and a dashboard decision with it.' That protects the sponsor's ability to decide without letting an unverified CPI authorise new scope.