The Benefits That Shrank

Principle in play: continued business justification. A PRINCE2 project must have a justified business case at all times, and the justification is re-confirmed at every stage boundary and whenever a forecast breach occurs - not once at approval. The benefit has fallen below the approved payback threshold, so the baseline justification no longer holds; that is a decision point, not a reporting detail.

Principle in play: manage by stages. Stage boundaries exist so senior management can decide whether to commit the next slice of resource. Deferring to closure removes every remaining decision point and converts a staged investment into an all-or-nothing bet.

Who decides. The executive is one voice of three. Continue/change/stop is a project board decision requiring the senior user (benefits realisation) and senior supplier (delivery options). 'It would look terrible' and money already spent are not valid justification - sunk cost is not a benefit.

What the project manager does: re-baseline the business case figures, raise an exception report at the earliest forecast, and put options (reduce scope to the £300k-saving channels, stop, or re-scope to close the gap) to the board in time for the next boundary.

The test is control, not practice. PRINCE2 does not mandate a number of management stages; combining stages is legitimate tailoring for small, low-risk, well understood work. The question is whether the board still retains a point at which it can review the business case and the next stage plan and decide whether to continue.

Apply three checks:

1. Does a senior decision point survive? Seven months with no board review, immediately after a 45% benefit reduction, removes the only mechanism by which the board could stop the project. That fails the test.
2. Is the discretion proportionate to risk? The risk profile has just increased (regulatory change, benefit uncertainty). Stage length should shorten as uncertainty rises, not lengthen.
3. Are stage boundary products still produced? An end stage report, benefits review and next stage plan are how the board discharges its accountability. If stages merge, those products must move to whatever boundary remains - they cannot simply be dropped.

Who decides: the board, not the project manager or the executive alone. Any agreed tailoring is recorded in the project initiation documentation (tailoring approach) and revisited at the next boundary.

Manage by exception cuts both ways. The project manager may not exceed delegated tolerances without authority, and may not silently absorb a change of this significance into 'business as usual'. But the project manager also cannot overrule the executive.

The correct sequence:

• Record and quantify. Log the forecast in the daily log and issues register with the evidence behind it, and update the business case so the variance is visible in writing.
• Escalate formally. Issue an exception report addressing the board through the executive, setting out cause, effect and options with a recommendation.
• Use project assurance. Project assurance is appointed by and reports to the board, independent of the project manager. Where the executive blocks escalation, informing project assurance - and, if necessary, the senior user and senior supplier - is the sanctioned route to put the issue before the whole board.
• Do not fabricate agreement. The project manager should not minute a board decision that was never taken, nor re-plan to make the numbers look tolerable.

If the board, properly informed, decides to continue, that is a valid decision - but it must be recorded, the business case re-baselined, and the rationale captured in the lessons log.