Alexander Pope wrote ‘To err is human; to forgive, divine,' pointing out everyone makes mistakes. Imperfection is part of being human, but accepting and forgiving those mistakes would elevate us to a higher plane of virtue. Yet a harder question lurks behind the saying: when did it become acceptable to repeat the same errors and simply assume forgiveness will follow?
That question matters in modern project delivery. Multidisciplinary teams jump from sprint to sprint, shifting scarce talent on to the next burning platform before the post-implementation dust has even settled. The UK's National Audit Office has warned only 8% of major Government projects are robustly evaluated and that ‘limited consequences for failing to evaluate' mean departments keep recycling identical problems. In effect, many public and private organisations have normalised Pope's ‘to err' half of the maxim, while quietly dropping the higher bar of learning or forgiving.
Enter the learning account (often called a lessons learnt review). Far more than a bureaucratic box-ticking exercise, a learning account is a deliberate, documented stocktake of what went well, what went badly and – crucially - what must change next time. Cabinet Office guidance lays out a four-stage lessons management framework: lesson identification; lesson prioritisation; lesson implementation; and embedding change. The message is explicit: capture observations; analyse and validate them; turn the most important into funded, time-bound actions; and keep checking until the new practice is truly ‘the way we do things round here'.
The cost of ignoring that message is written in the scorched remains of the Space Shuttle Columbia. Investigators found NASA's culture treated earlier foam-strike ‘near misses' where protective heat shields fell off the external tanks, as acceptable variance. Crucial knowledge stayed trapped in silos and dissenting engineers were unheard. After the 2003 disaster, the agency rebuilt itself as a ‘learning organisation', embedding technical authorities, mandatory post-flight reviews and open knowledge-sharing forums. Columbia shows lessons learnt exercises are not a luxury but a line of defence against catastrophe.
Why learning accounts work
- Institutional memory. Staff churn is inevitable; codified insights mean discoveries survive beyond individual careers.
- Risk reduction. Patterns of failure surface only when incidents are aggregated and interrogated.
- Psychological safety. A consistent, blame-light forum for reflection signals that speaking up is valued.
- Continuous improvement. Teams that inspect and adapt every cycle outperform those that rely on occasional heroics.
Making the practice meaningful
Yet many reviews wither and the opportunities to learn are lost. Common pitfalls include:
- Superficiality. A perfunctory review of past projects squeezed into the last half-hour of a project kickoff event.
- No owner, no budget. Recommendations die in spreadsheets because nobody is tasked (or funded) to act.
- Forgetting to check back. Actions are declared ‘closed' without evidence that anything actually changed.
The Cabinet Office framework offers practical advice: assign senior ‘lesson owners', insist on SMART actions and treat embedding change as a long-tail workstream rather than an afterthought.
Leadership behaviours that tip the balance
- Visible participation. When sponsors admit their own missteps, psychological safety soars.
- Learning velocity over blame velocity. Focus on how fast the team discovers and mitigates root causes.
- Forgive, but don't forget. Forgiveness should clear emotional decks, not wipe institutional memory.
From ‘when will we ever learn?' to ‘we are learning'
Pope called forgiveness divine; Wilde rebranded error as experience. Our task in 2025 is to connect the two, converting inevitable error into communal experience and ‘institutionalised forgiveness' that takes the form of robust, shared knowledge. Learning accounts operationalise that conversion. They answer the uncomfortable question implicit in those literary quips: it is not acceptable to keep committing the same mistakes and presuming absolution. Instead, organisations earn absolution by proving they have changed.
When project teams bring down the curtain on their latest achievement, they should not merely toast the handover; they should open the learning account, credit the hard-won insights and debit the risks for next time. Only then can we move from the rhetoric of ‘continuous improvement' to the reality: delivering the next initiative a little better than the last. Continuous improvement, after all, is simply structured forgiveness married to actionable memory - a union as human and as necessary as the errors that make it so.
In short, if you want innovation without déjà vu, make the learning account as non-negotiable as the go/no-go gate. Erring may remain human, but learning is what makes an organisation a little bit divine.
