Skip to content
Projects Asia-Pacific

Risk & uncertainty

Five questions before approving a major project

Five practical questions for sponsors, boards and executives testing a major project before approval.

By Jim Parke |

A major project can arrive at an approval meeting with an impressive business case, a detailed timetable and a compelling account of the benefits to come. Those documents matter. They do not, by themselves, tell a sponsor whether the organisation is ready to make the commitment being requested.

Approval is a decision made under uncertainty. The problem may be well established while the proposed solution remains immature. A cost estimate may be precise without being reliable. Benefits may be attractive without anyone having accepted responsibility for delivering them. The sponsor’s task is to test those distinctions before a preferred option becomes too difficult to reconsider.

The following five questions are a practical starting point for boards, senior executives and public-sector sponsors. They apply beyond infrastructure: major technology, organisational change and service-transformation programmes face many of the same approval challenges.

1. What problem are we solving, and what alternatives remain open?

Large projects can acquire momentum long before the final investment decision. A proposed building, platform or transport corridor becomes the project’s identity. Discussion then turns to how to deliver it, while the question of whether it is the best response receives less attention.

The sponsor should be able to state the underlying need without naming the preferred solution. What happens under business as usual? What is the do minimum option that meets the essential objective without unnecessary features? Can the proposal be staged, tested or combined with changes to existing services? If the alternatives have been dismissed, the reasons should be visible and capable of challenge.

This is not an invitation to delay every decision. It is a way to keep the investment case tied to the outcome sought. A project that is well managed but addresses the wrong problem is still a poor investment.

2. What would the forecast look like from outside the project team?

The people designing a project know its detail. They may also become attached to its success. An approval paper should therefore show more than the team’s preferred estimate of cost, duration and benefits.

Ask what happened on comparable projects. How often did similar undertakings meet their original budgets and completion dates? Which costs were omitted from early estimates? What happened to benefits after opening or implementation? This outside-view practice is often called reference class forecasting. It will never produce an exact answer for a unique project, but it can expose whether the forecast assumes an unusually favourable outcome without explaining why.

For UK public-sector appraisals, HM Treasury’s 2026 Green Book requires explicit adjustments for optimism bias at the outset. It directs practitioners to start with the originating organisation’s historical forecast errors, consider similar proposals and use generic adjustment values when the public body lacks its own evidence. The useful principle travels beyond that appraisal system: ask, “What was our forecast error last time?”, test the new forecast against relevant experience and explain the differences. A contingency is not a substitute for this exercise if nobody can explain what it covers or how it was calculated.

Sponsors should also ask to see a range, not just a single number, and understand where the proposal sits against comparable outcomes. Australian transport sponsors can look to the Australian Transport Assessment and Planning Guidelines for a national appraisal framework, while following the requirements of their own jurisdiction.

3. Which assumptions could change the investment decision?

Every business case contains assumptions. The critical ones deserve more prominence than a long register can give them. They are the assumptions that, if wrong, could change the preferred option, the scale of the investment or the decision to proceed at all.

Examples include future demand, access to land, the availability of specialist labour, an unresolved regulatory approval, the readiness of a partner organisation or the ability of users to adopt a new service. Each needs an owner, a basis in evidence and a date by which uncertainty can be reduced. Which assumptions drive the forecast range? What would have to be true for the low-cost or early-completion scenario to occur? A sponsor should be able to distinguish what has been independently tested from what is still a working proposition.

The next question is the switching value: how far would a key assumption have to move before the preferred option ceased to offer value for money or another option became preferable? HM Treasury’s Green Book calls for these values to be calculated for the preferred option. In plain terms: “How far would demand have to fall before we would choose differently?” That question tells a sponsor more than an unqualified assurance that the business case is “robust”.

This distinction is especially useful at staged approvals. An early decision may properly fund design, consultation or investigation. It should not silently authorise irreversible delivery commitments while decisive assumptions remain unresolved. State clearly what the current approval permits, what it does not permit, and what evidence the next decision will require.

4. Who will deliver and measure the benefits?

Benefits are often described at approval as though they will follow automatically from completing the asset or programme. Yet a new facility can open on time without delivering its intended service improvement. A digital system can go live without changing the practices that made the investment worthwhile.

For each material benefit, ask who owns it, what baseline will be used, when it should emerge and how it will be measured. Is the owner able to make the operational changes required? Has the cost of those changes been included? What happens if delivery finishes before the benefit can be observed?

The answer may sit outside the project team. A project manager can deliver an output, while an operating division, agency or executive must change processes, adopt the output and sustain the result. The Association for Project Management describes the sponsor as accountable for the realisation of specified benefits over time. Sponsorship should bridge that handover. Approval papers ought to identify the person or body that remains accountable for the benefits after the delivery team has left.

5. What would cause us to pause, change course or stop?

A credible approval is conditional. The sponsor should know how adverse information will reach decision-makers, who can challenge the prevailing view and which developments trigger a new decision. This is the purpose of governance and assurance: to improve decisions while there is still time to act.

Consider a project that encounters a major cost increase or a sharp reduction in expected demand. Who has authority to reappraise the business case? Will the board receive the changed forecast promptly, with the original baseline alongside it? Can independent reviewers speak directly to the sponsor? Is there a practical route for staff or contractors to raise concerns without those concerns being filtered through the people whose work is under review?

These questions should be answered before trouble arises. The Association for Project Management describes effective assurance as independent, objective and proportionate to the work, targeted where risks are greatest. A regular progress report is useful, but a series of green status indicators is not a decision framework. Sponsors need thresholds for escalation and genuine choices at each major gate: proceed, redesign, defer or stop. The decision record should explain both the evidence considered and the conditions attached to approval.

Approval is the beginning of active sponsorship

None of these questions promises certainty. Major projects require judgement, and reasonable people can disagree about the weight to give competing outcomes. The aim is a decision whose assumptions, alternatives and consequences are clear enough to be tested over time.

Before approving the next stage, a sponsor should be able to say: We know the problem we are trying to solve; we have challenged our forecast; we understand the assumptions that matter; someone owns the benefits; and we know when this decision must be revisited. If any part of that statement is missing, the right response may be a narrower approval that buys the evidence needed for the larger one.

That is not a lack of ambition. It is how ambitious projects earn the authority to proceed.

Further reading

Related articles

Start a conversation

Discuss what stronger project leadership could look like in your organisation.

Submit an enquiry