The Government Internal Audit Agency reviewed the Department for Transport’s lessons-learning capability in January 2026. It found the framework well-designed and clear. It also found that programme and project teams within DfT do not consistently apply the lessons they have identified. Two months later, the NAO published its report on Northern Powerhouse Rail. The recommendation for ensuring NPR learns from previous major programmes: regularly assess how lessons learned are being implemented.

The mechanism confirmed as non-functional is the mechanism being prescribed.

The exhibit

The NAO’s Northern Powerhouse Rail report, published 11 March 2026, notes that by February 2026 DfT had identified 11 lessons from previous rail programmes with 24 corresponding actions embedded into NPR planning. The lessons cover governance, skills and capabilities, and contract and financial management. The NAO acknowledges the GIAA’s January 2026 finding that DfT’s lessons framework is well-designed but inconsistently applied across the department. The NAO’s fourth recommendation to DfT is to ‘regularly assess how lessons learned from previous major programmes are being applied’ and to ‘report on how it is incorporating those lessons into the programme.’ £410 million has been spent on NPR development before construction has begun.

Why it is wrong

The GIAA confirmed in January 2026 that DfT produces documented lessons. Teams do not apply them. The question this finding raises is why. The answer the GIAA does not provide, and the NAO does not examine, is this: there is no consequence for a programme team that ignores a documented lesson. The lessons are captured centrally. The teams that build the next programme are under pressure to deliver on scope, time and budget. Demonstrating that the programme has internalised lessons from a predecessor is not a performance measure. It is not a gate criterion. Nobody’s career depends on it.

A monitoring framework produces a record of whether lessons are being applied. It does not produce application. The mechanism creates an audit trail of non-compliance. It does not create compliance. Recommending more monitoring of a mechanism that is already monitored and already documented as non-functional is not a prescription. It is the appearance of one.

HS2 generated its own lessons. DfT documented them. NPR is now being planned against 11 of those lessons, with 24 actions, by teams that the GIAA has confirmed do not consistently apply lessons. The NAO recommends more assessment. The problem is not assessment.

What a careful reader would write instead

If DfT’s teams consistently fail to apply documented lessons despite a well-designed framework, the failure is not in the framework. It is in the accountability structure around it. A functional lesson-application process has two components the DfT approach lacks: a named owner for each lesson accountable for its application in the next programme, and a consequence if it is not applied. Without those two elements, a lessons framework is a filing system. It records what went wrong. It does not change what happens next.

The NAO recommendation that DfT ‘regularly assess’ lesson implementation would be useful if the problem were that nobody was tracking it. The GIAA found the framework for doing so is already in place. The problem is enforcement, not visibility.

The punch

Every major UK infrastructure failure produces a review, a report, and a set of lessons. Crossrail did. HS2 did. The GIAA reviewed DfT’s mechanism for ensuring those lessons reach the teams building the next programme, confirmed it is well-designed, and confirmed it does not work. The procurement and programme management industry’s response to infrastructure failures has been to recommend better knowledge capture for two decades. The evidence from the department responsible for the UK’s largest rail programmes is that capture without accountability is paperwork. NPR is now accumulating its own paperwork. The reason teams are not applying the lessons is not that the lessons are hard to find. It is that nothing happens when they are ignored. The NAO’s prescription does not address that. It adds a layer of reporting on top of a process already confirmed as broken, and calls it a remedy.

Source: National Audit Office, Northern Powerhouse Rail, 11 March 2026, https://www.nao.org.uk/reports/northern-powerhouse-rail/