Who is allowed to say no?

Who Is Allowed to Say No?

Investment systems naturally produce advocacy. Sponsors know the upside, teams are rewarded for launch and sunk work creates a constituency for continuation. If no named role owns rejection, proposals accumulate until the portfolio exceeds capital, capacity or risk tolerance. A right to approve without a corresponding duty to decline is incomplete governance.

Define decision rights by exposure. Management can approve reversible experiments within a budget; larger or less reversible commitments require independent finance, risk and operational challenge; transformative bets belong with the board. Specify who recommends, who validates assumptions, who decides and who can pause after approval. Consultation should not blur accountability.

The decision-maker needs genuine alternatives: business as usual, do minimum, staged option and reallocation to another proposal. The 2026 UK Green Book requires broad option generation and retains business as usual as a benchmark, reducing the risk that a preselected asset is compared only with a weaker version of itself. Capital committees need the same discipline.

A credible �no� uses transparent criteria: strategic fit, incremental value, affordability, evidence, risk concentration and consumption of scarce capability. Record the reason and conditions for reconsideration. This protects teams from arbitrary veto while preventing negotiation from converting every failed threshold into a special exception.

Rejection is not the only control. The accountable authority should reduce scope, require an experiment, sequence a dependency or stop a funded project when its forward case changes. Track approval quality through forecast error, stopped capital and portfolio outcomes�not approval speed alone. Strong governance gives someone both the information and institutional permission to protect the next-best use of resources.