Program Budget and Capacity Planning

A budget can look healthy while the program has no realistic route to its next milestone. A vendor invoice is late, so actual spend is low. The test environment is unavailable, so the team has not started the work that will consume the budget. Another program has already booked the compliance specialist needed for approval.

Cost and capacity need to be read together. This guide uses invented figures to show how to do that; none are reported project results. The approach is a planning aid, to be adapted to the organization's finance and approval rules.

Build the cost baseline from the work

Break the program into workstreams or deliverables that have owners and acceptance criteria. For each, estimate internal effort, external services, environments, equipment, testing, training and transition. Decide which costs are charged to the program and which remain in operating budgets. Keep the treatment consistent so the investment is not made to look cheaper by moving costs elsewhere.

Attach timing and evidence to the estimate. A vendor quotation, an estimate based on a similar task and an untested assumption do not have equal confidence. Record the range and what needs to happen to narrow it. Include procurement lead times and partner commitments, not only development effort.

APM's Competence Framework treats budgeting and cost control as establishing estimates and budgets and tracking actual and forecast costs. The operating lesson is to keep the forecast current rather than treating the approved budget as a prediction that cannot change.

Keep actuals, commitments and forecast distinct

Actual cost is cost already recognized under the finance team's policy. An unpaid purchase order can still commit the business to future spending. Forecast-to-complete is the latest estimate for remaining work, including relevant outstanding commitments.

Use one cutoff date and agree how accruals are treated. Do not add outstanding commitments twice if they are already included in the forecast-to-complete. The following simplified example uses thousands of dollars and assumes the internal labor cost is charged consistently.

Here the forecast total is $37,000 actual plus $83,000 remaining, or $120,000. It is $10,000 over the approved $110,000 delivery baseline. Adding the $14,000 commitment again would overstate the forecast. Comparing $37,000 spent with $110,000 budget and declaring a saving would be equally misleading.

Assume a separately controlled $15,000 contingency allowance is also authorized. The sponsor may approve a $10,000 draw if the change qualifies under the agreed rules. That reduces the remaining allowance to $5,000; it does not make the overrun disappear or remove the need to investigate it. Show the original baseline, approved transfers, current baseline and remaining exposure separately.

Explain why the forecast changed

Separate scope changes, rate changes, rework, additional operating support and estimate errors. For each material movement, show the cause, current estimate range and decision needed. A forecast is useful before a supplier has invoiced or all implementation detail is known.

If the cost is unavoidable, say which commitment makes it so. If it is avoidable, show the options and their consequences for benefits and risk. Stop-work decisions should consider remaining costs and expected future value, rather than proceeding simply because money has already been spent.

Use the business-case guide to review whether the investment still makes sense. The cost report should inform that review, not replace it.

For a worked movement table and separate sponsor decisions, see explaining a changed programme forecast.

Put scarce skills on the schedule

A headcount total does not show whether the right person is available at the right time. Ask workstream leads to estimate demand by skill and period. Deduct leave, support duties and existing commitments before accepting available capacity. Avoid assuming every hour can be assigned to planned change.

APM's resource-management guidance links allocation and aggregated demand to scheduling. Apply that thinking to compliance reviewers, finance acceptance, partner certification and operations training as well as engineers.

For illustration, suppose next month's usable change capacity is 30 engineering days and eight compliance days, after existing-service work has been reserved.

All three require 37 engineering and 11 compliance days. The portfolio is not feasible in that period. A and C together also exceed compliance capacity, even though their engineering demand fits. A and B fit both constraints at 22 engineering and seven compliance days. That is one feasible combination, not proof it has the best commercial value.

The sponsor could authorize A and B while commissioning a smaller discovery task for C only if that task fits the remaining capacity and has a useful deliverable. Alternatives include changing a deadline through the authorized route, reducing scope or adding genuinely qualified capacity. “Start everything and work harder” is not a resource plan.

Make the portfolio decision explicit

APM describes portfolio management in terms of strategic alignment and delivery capacity. A practical decision record should show:

• verified obligations and the consequences of missing them;
• the contribution of optional work to the business case;
• scarce skills, partner windows and dependencies;
• feasible combinations, not just individually attractive projects;
• work to stop, defer or narrow, with the owner of the consequence;
• the trigger for reconsidering the allocation.

A mandatory label needs evidence and an accountable owner. A scoring model should not hide that an alleged deadline is only a preference. Equally, a high optional-project score cannot override a genuine obligation without authorization.

Connect schedule, cost and change control

Moving a test window can extend vendor support, delay revenue and consume a specialist needed elsewhere. Assess those effects together. Preserve the approved baseline, publish the current forecast and seek approval for changes outside delegated tolerance.

For the illustrative table, if Initiative A needs two extra compliance days, A plus B now needs nine, above the available eight. The decision is not just an amended date for A. Re-sequence B's approval work, obtain another qualified reviewer or revise the portfolio commitment. Update the cost forecast for any associated delay or support extension.

Do not conceal variance by repeatedly resetting the baseline. Record who approved the change, why, and which version replaced it. Delivery teams need both the current working plan and a trace of the commitment it changed.

What to bring to a review

Bring a current forecast, the assumptions that moved, a capacity view by bottleneck skill and the decision required. Explain the remaining exposure, including unresolved estimates. Ask finance to reconcile actuals and commitments, and ask receiving operations teams to confirm transition effort and ongoing support funding.

A forecast can be uncertain and still useful. It becomes unreliable when the uncertainty is hidden or when work has been allocated to capacity that does not exist.

FAQ

Does low actual spend mean a program is under budget? No. Check accepted progress, outstanding commitments and the remaining forecast. Delayed work or delayed invoices can make actual spend look favorable while the final forecast is worse.

Should contingency be a fixed percentage? Not by default. Establish an allowance and approval rules suited to identified uncertainty and the organization's finance policy. Avoid treating a generic percentage as evidence that every risk is covered.

Can separate projects each be feasible while the portfolio is not? Yes. Shared reviewers, engineers, environments or partner windows can be over-assigned. Aggregate demand before accepting the combined plan.