The Sponsor's 'Small' Addition

Do not implement, and do not refuse. Log and size. The request was made verbally in a steering meeting; the first action is to raise a change request the same day with the sponsor as requester. Without a logged CR there is no audit trail and no defensible basis for choosing between the new screens and the mandatory payroll change.

Size it against real capacity, not against optimism. 180 person-days in ten weeks. If the delivery team is about 14 FTE, remaining capacity is roughly 14 x 50 = 700 person-days, so the request is about 26% additional work in the remaining window. A team at 95% utilisation has no 180 free days: the request can only be traded, not absorbed. State that plainly in the impact assessment, with the number, because 'we are busy' loses and '26% more work in ten weeks' wins.

Set the priority order explicitly. The statutory payroll reporting change is mandatory and date-driven; it must be funded first and, if it cannot wait nine days for the board, routed through the emergency change procedure with retrospective ratification. Self-service and mobile leave approval are discretionary benefits and sit behind it in the queue regardless of who asked.

Bring options, not a verdict. Three honest choices for the board: defer to release two; part-deliver (manager self-service now, mobile approval later); or hold the date and descope 180 person-days of baselined requirements, named individually. A sponsor who has previously overridden the board responds to a menu of costs far better than to a 'no'.

Separate authority from consequence. The sponsor can approve the change; the sponsor cannot approve it for free. Your job is to make the trade-off explicit, recorded and owned, then execute whatever decision is actually taken.

1. One-page written impact statement, circulated before the board. 180 person-days; the date impact (go-live moves three to four weeks, or baselined scope is removed); the 95% utilisation and 3% schedule reserve consumed; the statutory change ahead of it in the queue. Written, dated, addressed to the sponsor and copied to the PMO.
2. Decision menu. 'Approve with go-live moved to date X', 'approve with items A and B removed from this release', or 'defer to release two'. Sponsors refuse a refusal; they rarely refuse a menu.
3. If overridden, record it properly. Update the change log, risk register and decision log with the sponsor's name, the accepted date or scope impact, and the residual risk, and confirm by email. This is not point-scoring: when the date slips in week nine, the record shows what was known and who chose it, and the team is not left carrying an undocumented commitment.

Where to escalate. If the override would breach the statutory reporting obligation or a contractual commitment, escalate once to the portfolio owner or PMO in factual terms, because compliance and contractual exposure typically sit outside the sponsor's delegated authority.

What to avoid. Absorbing the work quietly to protect the relationship. The team will absorb it through overtime and test compression, and you will own the failure when the statutory change lands late.

Reserves are held against identified risks, not against scope additions. The 8% contingency exists so the project can absorb the risks already in the register - data quality, interface readiness, key-person loss. Spending it on manager self-service converts a risk reserve into a scope budget, and the project then enters go-live with a statutory change pending and no hedge.

The 3% schedule reserve is smaller than it sounds. On ten remaining weeks, 3% is roughly one and a half days. It cannot cover the three to four week impact of the requested features, and it is already partly consumed. Quoting it as available capacity would be misleading.

Decision rules to state to the board:

• A scope change is funded from new money or from descoping an equal amount of baseline work, never from contingency.
• If contingency is used for anything, the register must show which risk retired, and the reserve must be replenished or the exposure explicitly accepted by the sponsor.
• At 95% utilisation there is no capacity slack, so any reserve-consuming risk that materialises hits the date directly rather than being absorbed by float.

The uncomfortable point worth saying once. A team at 95% utilisation with two reserves partly consumed and a mandatory compliance change inbound is not a project with room for discretionary scope. If the sponsor still wants the features, the correct conversation is about which baseline requirements are removed, not about which reserve is raided.