Earned Value and an Unrealistic Recovery Demand
A project manager is six months into a twelve-month core network upgrade with a budget at completion (BAC) of $3.2 million. At the end of month 6 the control account reports planned value of $1.60 million, earned value of $1.28 million and actual cost of $1.52 million. The critical path runs through a data-centre cutover that has already consumed its 10 days of float. The sponsor has just learned that a regulatory filing depends on the cutover and demands that the original finish date be held 'at any cost'. Overtime is available but is charged at 1.5 times the standard rate; two of the fourteen team members are shared with another programme and can give at most 50 percent time until month 9. The change control board has approved no scope changes. The PM must present a recovery plan at next week's steering committee.
1. Calculate SPI, CPI, EAC on a CPI basis and TCPI to the original BAC, and explain what they mean for this project.
Read the worked answer
SPI = EV/PV = 1.28/1.60 = 0.80. CPI = EV/AC = 1.28/1.52 = 0.84. The project has earned 80 percent of the value planned for month 6 and is spending about $1.19 for every $1 of earned value. EAC (CPI-based) = BAC/CPI = 3.20/0.842 = approximately $3.80 million, an overrun at completion of about $0.60 million if the trend continues. TCPI to the original BAC = (BAC − EV)/(BAC − AC) = (3.20 − 1.28)/(3.20 − 1.52) = 1.92/1.68 = 1.14. The remaining work must therefore be delivered at 114 percent of the original cost efficiency-14 percent better than plan-while the team is currently performing at 84 percent. That is a very demanding target and should not be presented as a realistic base case without a specific, costed recovery plan. The PM should also test whether the variance is concentrated in a few work packages or is systemic, and should check the critical-path cutover, which has no float left. Do not simply extrapolate the CPI; establish the cause before committing to an EAC.
2. The sponsor demands the original date 'at any cost'. What options should the PM present, and what should the PM recommend?
Read the worked answer
'At any cost' is a direction, not an approved plan. The PM should present options with schedule, cost, risk and quality consequences: crash the cutover with overtime at 1.5 times the standard rate and additional resources, but note that the two shared staff are capped at 50 percent until month 9, so the critical path may not respond; fast-track by overlapping configuration and testing, which raises rework and quality risk; descope non-mandatory functionality to protect the regulatory dependency, which requires change control; re-baseline the finish date and manage the regulatory consequence openly; or accept a phased cutover with partial migration. The PM should recommend the option that meets the regulatory dependency at the lowest total risk, not the one that preserves the date on paper. The sponsor must decide explicitly, and the decision must be recorded. If the original date is retained, the PM should state the residual risk, the additional budget required and the contingency trigger points. 'Any cost' should be converted into a delegated budget limit and a formal change request rather than absorbed silently by the delivery team.
3. What should the PM communicate at the steering committee, and what should the PM avoid saying?
Read the worked answer
The PM should communicate a clear evidence-based position: current SPI and CPI, the CPI-based EAC as a forecast rather than a certainty, the fact that the cutover has consumed its float, the constraint on the two shared staff, and the options with their cost and risk. Use a forecast range where the data are uncertain, and distinguish facts from assumptions. State what decision is required from the committee, by when, and what the PM will do in the absence of a decision. Avoid saying 'the project is fine', 'we will catch up' without a mechanism, or 'the team must work harder'; those statements hide the forecast and transfer risk to people who cannot resolve it. Avoid blaming individuals for the variance before the cause is established. Also avoid committing to 'any cost' on the sponsor's behalf without an approved budget limit. A short, honest bad-news message with options preserves credibility better than an optimistic message that collapses at month 9.