35 Moving from programme delivery into day-to-day operations is a complex task and bodies must take particular care to manage dependencies at this stage. Some major programmes carry a high risk of disrupting existing services, especially when they are expected to lead to transformational change.
36 Nonetheless, bodies often underestimate the complexity of bringing a programme into use and may not begin planning for operations until the programme is nearing completion. For example, the Thameslink programme aimed to increase capacity and relieve crowding on rail services in London and the South East. It represented a substantial change in how the railway functions in this region, involving new technology, new ways of managing passengers at stations and new maintenance arrangements. However, the collective rail industry, including the Department for Transport as programme sponsor and Network Rail as network operator, did not start planning sufficiently early enough to manage the transition to bring the enhanced services into use. In May 2018, there was severe disruption to the rail network in that region, due to deficiencies in planning the introduction of the Thameslink services.11,12
37 A flawed understanding of how a programme will be used in practice can lead to gaps in its scope, requiring costly changes later on, and also to bodies making programme decisions with potentially far-reaching unintended consequences. If bodies do not plan effectively and early enough, the limits of what can be in place on day one may suddenly become apparent, and then need to be addressed rapidly to meet the programme schedule.
38 The risks around moving into day to day operations are made worse by the temptation to squeeze the time allotted for operational and systems testing to make up for delays that have occurred earlier in the programme, as we explored earlier on being realistic about the schedule (paragraphs 21 to 25). Furthermore, a new or enhanced service may require significant changes to ways of working and/or depend on additional capability in existing services to support it.
39 For example, as we reported in Carrier Strike - Preparing for Deployment, Carrier Strike relies on the sustained availability of munitions and stores, such as ammunition and food. However, the Ministry of Defence has been slow to develop the support ships needed to supply the carriers and currently has only one suitable ship. It has long been aware that this will restrict the operational freedom of Carrier Strike but has not yet developed a solution. In November 2019, the Department stopped the competition to build three new support ships because of concerns about value for money. It now believes that this will delay the introduction of new ships by between 18 and 36 months, making it uncertain that the first new support ship will be operational before the existing one leaves service in 2028.13
40 There is often an ambition to introduce a new service in a 'Big Bang' improvement, rather than in smaller stages that can be tested, refined, and embedded more carefully. For a staged approach to work, however, bodies must plan each stage in advance, considering what they hope to learn from each stage and what risks they are seeking to mitigate. They should not assume that the use of stages alone will solve any problems that a programme may encounter. For example, in introducing the Personal Independence Payment, the Department for Work & Pensions (DWP) did not identify the most significant risks to test, and did not allow enough time between stages to assess performance across the full benefit process before it increased volumes. Backlogs developed early and the Department did not have sufficient time to resolve problems before the next stage of implementation.14 We have also seen cases where introducing stages at the last minute has caused problems because bodies have not fully thought through the consequences of rapidly changing planned arrangements.
41 While timely planning can avert potential problems, bodies should also expect that unknown issues will emerge, and should allow time for these to be discovered and managed. For this reason, bodies must continue to have contingency arrangements in place in case of service failure, regardless of the approach taken.
_____________________________________________________________________
11 Comptroller and Auditor General, Update on the Thameslink Programme, National Audit Office, HC 413, Session 2017-2019, November 2017.
12 Office of Rail and Road, Independent inquiry into the timetable disruption May 2018, December 2018. Available at: www.orr.gov.uk/sites/default/files/om/inquiry-into-may-2018-timetable-disruption-december-2018-report-grayscale.pdf
13 Comptroller and Auditor General, Carrier Strike - preparing for deployment, National Audit Office, HC 374, Session 2019-2021, June 2020.
14 Comptroller and Auditor General, Welfare Reform - Lessons Learned, National Audit Office, HC 77, Session 2015-16, May 2015