The Untested Hour: Giving Operational Rehearsal a Budget Line in UK Delivery

Most UK delivery plans are financially complete until the moment the organisation has to operate what has been built. Build is costed, test is costed, launch communications are costed. The rehearsal — the dry run in which people, handoffs and fallback positions are exercised together — is often left to goodwill, overtime and the hope that nothing unusual happens in the first week. That is not a control position. It is a funding gap with a delayed invoice.

Testing the system is not the same as rehearsing the organisation

A test environment can confirm that a transaction posts correctly. It rarely confirms that a team can absorb a new queue at month-end, that a supplier handoff works when the usual supervisor is on leave, or that a manual fallback can be operated for half a day without creating a backlog. Those are operational questions, and they carry operational costs: staff time, backfill, out-of-hours cover, temporary workarounds and the management attention needed to resolve what the rehearsal exposes.

When those costs are not named, they do not disappear. They are absorbed by teams already carrying delivery work, or they are deferred until go-live, where they appear as incident response. A rehearsal budget simply moves the cost to the point where it can still influence the outcome.

What the line should buy

A credible rehearsal budget is not a general contingency. It should be scoped against specific activities and released against a short plan. At a minimum, it should cover four things.

First, participant time and backfill. A dry run that depends on front-line staff volunteering around shifts is not a rehearsal; it is a sample. Backfill for operational participants should be costed in the same way as any other project resource.

Second, an end-to-end handoff exercise. This should follow a live-like case across team boundaries, including the points where ownership changes and where information is re-keyed or checked. The purpose is to find the quiet delays that no single team owns.

Third, a fallback rehearsal. Every delivery plan has fallback positions on paper. Few are operated before go-live. A short exercise in which the primary route is deliberately unavailable — a supplier interface, a system queue, a key approver — tests whether the fallback is usable, staffed and understood.

Fourth, the first 48 hours of live service. This is not a euphemism for hypercare. It is a funded rota, a named decision-maker, a route for operational issues and a plan for the first full business cycle. If the first 48 hours are funded only by seniority and goodwill, the organisation is relying on the same people who are already tired from delivery.

A scope, not a blank cheque

The budget should be owned by the delivery lead but released against a rehearsal scope agreed with the operational owner. That scope should name the scenarios to be exercised: peak volume, supplier unavailability, absence of a key person, a data quality exception and a handoff between two teams. It should also name the decisions the rehearsal is expected to inform, such as whether a manual workaround is viable, whether a queue needs additional cover, or whether a go-live date should move.

This is the difference between a rehearsal and a demonstration. A demonstration is designed to pass. A rehearsal is designed to surface a decision. The budget should be attached to the latter.

Sizing and releasing the line

There is no defensible percentage of project cost that applies to every delivery. Sizing should be built from participant days, backfill cost, out-of-hours cover and the time needed to capture and act on findings. A modest line, calculated in days rather than adjectives, is usually enough. The comparison is not between rehearsal and no rehearsal. It is between rehearsal and the cost of discovering the same issues during live service, when every hour carries operational and reputational weight.

Release should be conditional on evidence, not on a diary entry. Before the money is drawn, the programme should be able to show a rehearsal plan, a participant roster, a scenario list, a fallback test and a first-48-hours rota. After the rehearsal, it should show a findings log with named owners and dates. That evidence does not need to be elaborate. It needs to be sufficient for the operational owner to say that the organisation has been exercised, not merely informed.

The cultural point behind the line

Naming a rehearsal budget does more than allocate cost. It removes the assumption that operational readiness can be delivered through extra effort at the edge of the plan. It makes the dry run a managed activity with a scope, an owner and an end date. For UK organisations working within fixed financial years and constrained operational capacity, that discipline is not bureaucratic. It is a way of preventing the first week of live service from becoming the first real test.

The untested hour is rarely the hour anyone planned for. Funding the rehearsal does not guarantee a smooth launch, but it makes the likely points of failure visible while there is still time to act. That is a better use of budget than discovering them after the ribbon has been cut.