Green by Re-Baseline
You take over a ten-month project in month 7. Of twelve deliverables, six are complete. The remaining work is estimated at 5.5 months of the team's capacity. Earned value gives SPI 0.72 and CPI 0.83. The programme manager tells you the board 'has no appetite for bad news this quarter' and suggests moving the baseline end date out by two months and reporting amber. Two team members tell you privately that the original estimates were squeezed to win approval and that they were told so at the time. A regulatory go-live is fixed at month 10, with contractual penalties of EUR 25k per month of delay after it.
1. How do you respond to the request to re-baseline and report amber?
Read the worked answer
Distinguish a legitimate re-baseline from a cosmetic restatement, and refuse the second while offering the first.
A re-baseline is legitimate when scope or an external constraint has genuinely changed and the change is approved through governance. Here nothing has changed except the visibility of the variance - moving the date by two months does not move the work, and the project is 5.5 months from done with three months left.
My counter-proposal:
- Keep the baseline intact so the variance stays measurable.
- Submit a formal forecast revision through the change board with a stated cause (estimate quality at approval) and a documented recovery plan.
- Report red on schedule, amber on nothing, with the forecast date and the EUR 25k/month penalty exposure stated in numbers.
Then summarise the conversation to the programme manager in a short email. Not as a record against him - as confirmation of what was agreed, which is also the thing that protects him if the board later asks why the change was or wasn't made.
2. With three months to the fixed go-live and 5.5 months of work, what options do you put to the board?
Read the worked answer
Three costed options, one decision, now.
- Compress: add people or overtime to deliver full scope by month 10. Cost the ramp-up, the productivity dip from onboarding, and the quality risk. Compare against EUR 25k per month of penalty - for many projects compression is cheaper than two months of penalty plus the commercial damage, but only if the critical path can actually absorb more people.
- Descope to the regulatory minimum by month 10 and deliver the remainder in a second release. Identify which of the six open deliverables are actually required by the regulator - usually fewer than stakeholders assume. This is often the only option that holds the date at near-zero cost.
- Accept the penalty and hold scope and quality. Quantify: EUR 25k × months late, plus the contractual and relationship consequences.
What must not happen is option 3 being selected by default in month 9 because nobody chose in month 7. The decision is the board's; the duty to put a dated decision in front of them is mine.
3. What do you do with the team members' disclosure about the pressured estimates?
Read the worked answer
Treat it as a governance finding, not a disciplinary matter - and protect the source.
- Corroborate, don't interrogate. Check whether the estimate compression is visible in the approval record (estimate revisions, review notes, decision dates) rather than requiring individuals to testify.
- Use it as the documented cause in the forecast revision: "estimates were reduced during the approval process without corresponding scope reduction." That is a process finding and it is the honest explanation for a 0.72 SPI at month 7.
- Protect the sources. Do not name them in a board paper. If I am asked directly who told me, I say the finding is corroborated by the approval record - which it should be, or I would not be relying on it.
- Feed it into lessons learned and the estimating process, so the next business case is not built the same way.
What I would not do is use the disclosure as a shield - "the team lied at the start" - because the estimates were signed off by the governance that is now asking for a cosmetic baseline, and that includes the programme manager.