Finance Never Believes Programme Benefits Because the Business Case Hides the Uncertainty
A benefits case is not a promise that the future has been measured; it is a controlled statement of what must become true for the investment to remain justified.
The Meeting Where Two Kinds of Truth Collide
The business case has been in circulation for three weeks. The programme director believes it is conservative: £18 million of benefit against £11 million of investment, with a contingency on cost and a cautious adoption curve. The finance director sees something else. Only £3 million will leave a budget. The rest depends on time saved, better purchasing, fewer errors and decisions that are supposed to become faster once a new information system is in place.
Both are intelligent. Both are acting responsibly. Yet by the end of the meeting each believes the other has failed to understand value.
This argument recurs because finance and programme teams are not merely using different vocabulary. They are answering different questions. Finance asks, “What will change in the accounts, when, and who will answer if it does not?” The programme asks, “What improvement makes this investment worth attempting, including value the accounts cannot capture?” The first is concerned with proof and stewardship. The second is concerned with possibility and change. A business case asks them to occupy the same page without giving them a shared discipline for doing so.
In 2016, the tension is especially sharp. Years of constrained spending have made chief financial officers rightly suspicious of optimistic cases. At the same time, organisations are investing in digital channels, integrated information, shared services and new operating models whose value cannot be reduced to headcount alone. The old bargain — approve the project on a large forecast, then look for the benefits after delivery — has exhausted its credibility.
The roots of the dispute lie deeper than poor templates. They lie in what each profession has learned to fear.
Finance Remembers the Benefits That Never Reached the Ledger
A finance team sees the accumulated residue of past promises. It sees programmes that claimed efficiency but left the cost base unchanged; systems that saved thousands of staff hours but required the same establishment; consolidations that forecast procurement leverage while local buying continued; service improvements that were declared complete without a stable baseline.
The scepticism is not cultural resistance. It is institutional memory.
Take a composite transformation intended to combine six regional service operations. The approved case promises £12 million over three years. Of that, £5 million is supplier and property reduction, £4 million is staff capacity, and £3 million is improved quality. Eighteen months after go-live, the supplier contracts have been renegotiated and two leases surrendered. Finance can see £4.4 million. The programme reports that 310,000 hours of handling time have also been removed. Yet no posts have been taken from budgets, volumes have risen, temporary staff remain, and local managers have used the time differently.
The programme calls the capacity benefit real. Finance calls it unbanked. Both descriptions may be correct.
The fracture appears because the original business case treated unlike things as though they were equivalent. A lease surrender, an hour released and a reduction in customer complaints were all converted into pounds and added together. Once aggregated, they looked equally certain and equally available. They were neither.
The common currency of money creates an illusion of comparability when the underlying claims have different mechanisms, owners and degrees of certainty.
Financial scrutiny then arrives late, often as a challenge to a number rather than an inquiry into its mechanism. By that point the programme experiences finance as a gatekeeper: a function that accepted the strategic ambition in principle but rejects the arithmetic used to fund it. Finance, meanwhile, experiences the programme as a seller whose claims become more emphatic as the evidence becomes weaker.
Distrust is the predictable result of a process that asks one side to sell and the other to police.
Programmes Remember the Value Finance Could Not See
The programme side carries its own history. It has seen worthwhile changes delayed because only direct cash savings counted. It has watched mandatory controls, service resilience, customer access and management information treated as second-class benefits because their value could not be expressed cleanly in a discounted cash flow.
A financial test can become too narrow for the decision it is meant to support. If a new customer channel reduces avoidable visits, improves access outside office hours and creates cleaner demand information, the absence of an immediate budget reduction does not make the investment valueless. If a control improvement prevents recurring errors, its worth is not confined to the average cost of errors recorded last year. If common data gives executives a consistent view of performance, the result may matter before anyone can prove that a particular decision was better.
Programme leaders therefore learn to translate strategic and operational value into the language that secures approval. Hours receive salary rates. Better information receives a productivity percentage. Improved morale receives a retention assumption. Risks receive expected monetary values built from probabilities no one can substantiate.
This is often described as optimism bias, and sometimes it is. More often it is a rational response to a funding system that recognises a benefit only after it has been made to look financial.
The strongest defence of finance is that capital allocation requires comparison. An executive team cannot choose between twenty proposals using adjectives. Numbers expose assumptions, enable challenge and force sponsors to state scale. That defence is serious. Without common measures, persuasive sponsors can dominate and strategic language can excuse almost anything.
But the programme objection is equally serious: a false common measure does not improve the decision. It merely moves judgement into assumptions where it is harder to see.
The Business Case Became a Treaty, Then a Sales Document
The modern business case has inherited several purposes that do not sit comfortably together. It is expected to justify strategy, secure capital, define scope, predict delivery cost, quantify benefits, allocate accountability and later serve as the baseline for assurance. These purposes have different standards of evidence.
At approval, uncertainty is unavoidable. The design may be incomplete, supplier pricing provisional and adoption dependent on future decisions. Yet governance often rewards a single number. A range looks indecisive. A contingent benefit looks weak. An explicit unknown looks like poor preparation. Sponsors learn that certainty travels through committees more easily than honesty.
The paper therefore becomes a treaty between ambition and accounting, negotiated under pressure to make the ratio work. Once approved, the treaty hardens. Assumptions that were provisional become commitments. A planning estimate becomes a target. A target becomes a forecast. Later, the same number is reported as though it had always been a measurable outcome.
A typical chain might look like this:
- A proposal assumes 8 per cent less processing effort after standardisation.
- The effort is multiplied by the full salary cost of 900 staff.
- The resulting £4.6 million is entered as an annual benefit.
- The operating budget is not reduced because service demand is expected to grow.
- The programme nevertheless reports £4.6 million “realised” when the new process goes live.
Nothing in the chain requires dishonesty. Each step answers a different need. Together they produce a claim that finance cannot reconcile and operations cannot own.
The deeper problem is temporal. Finance wants a forecast before investment; genuine benefit evidence emerges after behaviour changes. Programmes bridge the interval with assumptions. That bridge is necessary. The mistake is pretending it is already solid ground.
A benefits case is not a promise that the future has been measured; it is a controlled statement of what must become true for the investment to remain justified.
What a Shared Language Would Have to Preserve
Bridging the divide does not mean teaching programme managers more finance or asking finance teams to become more sympathetic. Both help, but neither resolves the structural problem. A shared language must preserve the legitimate concerns on both sides.
It must preserve finance’s insistence on:
- Traceability: the route from intervention to financial or operational result can be followed.
- Additionality: the outcome would not occur without the investment.
- Timing: the benefit appears in a defined period, not indefinitely in the future.
- Ownership: someone with relevant authority accepts responsibility.
- Reconciliation: cash claims can be matched to budgets and accounts.
It must also preserve the programme’s insistence on:
- Strategic value: some investments matter because they enable a direction, not because they remove cost immediately.
- Operational reality: released capacity may absorb growth, improve service or avoid future expenditure.
- Uncertainty: early estimates should develop as design and evidence improve.
- Behaviour: technology and process do not create value unless people use them differently.
- System effects: benefits may cross organisational boundaries and resist attribution to one project.
The shared language begins by refusing to collapse these claims into one total.
| Benefit class | Evidence before approval | Evidence after delivery | Required owner |
|---|---|---|---|
| Cash release | Budget line, action and date | Reduced expenditure or avoided commitment | Budget holder |
| Capacity release | Time or volume baseline | Measured capacity and redeployment decision | Operational leader |
| Service improvement | Current service measure | Sustained change against baseline | Service owner |
| Risk reduction | Exposure and control weakness | Control performance and residual exposure | Risk owner |
| Strategic enablement | Defined decision or capability | Evidence that the capability is used | Executive sponsor |
This classification does more than improve reporting. It changes the conversation at approval. A £10 million case might contain £2 million cash release, £3 million capacity, £1 million cost avoidance, and several service and strategic outcomes that are deliberately not monetised. The case may still be compelling. It is simply no longer allowed to imply that £10 million will appear in the accounts.
Ownership Is Where the Languages Finally Meet
Finance and programmes can agree on definitions and still fail if ownership remains ceremonial. The phrase “benefit owner” appears in many governance packs, but the named person often lacks control over the budget, process, policy or workforce that must change.
Suppose a programme introduces a common purchasing system and forecasts lower prices through aggregated demand. The programme can deliver the system, supplier data and category reports. It cannot force operating units to combine orders, retire local arrangements or accept standard specifications. Those decisions sit with business leaders. If they do not make them, the benefit is not delayed by the programme; it has been declined by the organisation.
This is where finance scrutiny should move upstream. Instead of challenging only the percentage saving, finance should ask whether the owner has the authority and plan to change purchasing behaviour. Instead of requiring the programme to “realise” a benefit after handover, the portfolio board should decide which operational commitments are conditions of continued investment.
The practical sequence is simple, though rarely comfortable:
- State the outcome without money first.
- Describe the causal chain from deliverable to changed behaviour to result.
- Classify the benefit before valuing it.
- Name the operational and financial decisions required.
- Assign each decision to an owner with authority.
- Set the evidence and date for revisiting the claim.
- Adjust, stop or reinvest when the evidence changes.
The important move is the last one. Most business cases are approved once and defended thereafter. A credible benefits case is a living investment thesis. Its assumptions should become more accurate as the organisation learns. Reducing a forecast is not failure if the reduction prevents further misallocation. Increasing investment can be justified if evidence strengthens. The discipline is not adherence to the original number; it is fidelity to the decision.
Neither Side Can Own This Alone
There is a temptation to solve the problem by giving benefits governance to finance. That would increase rigour, but it risks narrowing value to what can be reconciled. Giving it to the programme preserves the change logic, but leaves the seller marking its own promises. Assigning it to a portfolio office can create consistency, but not authority over budgets or operations.
The work must be jointly held and distinctly accountable. Finance owns the integrity of financial classification and reconciliation. Programme leadership owns the intervention logic and delivery evidence. Operations owns behaviour change and service outcomes. The sponsor owns the continuing investment judgement. The portfolio board owns the decision to continue, reshape or stop.
That distribution may feel less tidy than appointing a single benefits function. It is more truthful because value itself is distributed.
The roots of the CFO’s disbelief are therefore not hostility to programmes. They are years of cases that converted uncertainty into precision, capacity into cash and aspiration into commitment. The roots of the programme leader’s frustration are not indifference to stewardship. They are years of being asked to justify strategic change through measures designed for financial transactions.
The divide closes when both sides stop demanding that one language defeat the other. Finance must accept that not all value reaches the ledger. Programmes must accept that value not reaching the ledger still requires evidence, ownership and consequences. Between those positions lies a business case that can do what it was always meant to do: support a decision under uncertainty without pretending the uncertainty has disappeared.