Cutting training from 10 days to 2 days is one of those decisions that looks brilliant on a spreadsheet and disastrous in practice — a classic example of mistaking visible cost for actual cost. The £15m saving is real and immediate; the cost of cutting training is real but mostly delayed and invisible until 3-5 years later, by which time the damage is done. The CEO's logic ('they can learn on the job') sounds reasonable but is fundamentally wrong for a complex regulated business.
The visible saving — what the spreadsheet shows
- £15m direct cost saving (training providers, materials, room hire, certifications).
- Productivity recovery: 8 days × workers actively in roles instead of training = more output.
- Faster onboarding (fewer mandatory training days).
- One-time positive impact on CEO's first-year P&L.
These numbers ARE real. The CEO is not making them up.
The invisible cost — what the spreadsheet doesn't show
1. Compliance failures will rise. Banking is heavily regulated. Money laundering rules, financial advice standards, market conduct, GDPR, cybersecurity — all require ongoing training to maintain compliance. Cut training and the firm will, over 1-3 years, fail more compliance audits, face larger fines, and possibly lose its licence to operate in some areas. Single regulatory failures regularly cost banks £50m-£500m+. The £15m saving is a rounding error against this.
2. Customer-facing errors will rise. Banks make decisions on lending, fraud, mortgage approvals, investment advice — all requiring up-to-date expertise. Under-trained staff make more errors, leading to customer complaints, redress payments (which can run to billions in the UK banking industry), litigation, and brand damage.
3. Skill atrophy. Banking products, technologies, regulations evolve constantly. Staff who don't train fall behind — and the firm with under-trained staff becomes uncompetitive. The decline is gradual but compounding.
4. Retention damage. Investment in development is a major motivator. Staff who feel the firm has stopped investing in them disengage, then leave. Replacement cost (3-6 months salary per departure) wipes out training cost savings within 2-3 years.
5. Cultural signal. Cutting training signals that staff development is not a priority. The signal demoralises remaining staff, weakens recruitment ('this firm doesn't develop you'), and accelerates the very turnover the firm can least afford in a tight labour market.
6. Strategic capability erosion. The bank competes on judgement, expertise and trust. Cutting investment in those exact capabilities undermines the source of competitive advantage. Tech firms invest 10-20% of revenue in R&D + training; banks that try to compete on cost-cutting will lose to those that invest.
Why the CEO's argument is wrong
The 'learn on the job' argument fails for three reasons:
- Banking work is complex and consequential. Errors aren't recoverable like errors in simple roles. A trader who hasn't been trained on derivatives risk can lose millions in minutes.
- Banking is regulated. Many topics legally require off-the-job training to recognised standards. 'Learn it on the job' is not a defensible position to a regulator.
- Knowledge tacit on-the-job is outdated. The senior people that juniors would learn from are themselves operating on knowledge from 5-10 years ago. Without structured external training, the whole organisation drifts.
The alternative the CEO should consider
Rather than cut training to 2 days, the bank should:
Audit training spend. Some 10-day-per-year training programmes ARE genuinely low-value (compliance modules nobody remembers, generic 'leadership' courses, training of leavers). Cutting selectively may save 30-40% of training cost while preserving the genuinely valuable training.
Shift mix. Move some training to more efficient channels (online, self-paced, micro-learning) — same outcome at lower cost than in-person workshops. Aim for 6-7 effective training days, not 10 inefficient ones.
Measure ROI. Most banks don't measure which training actually changes behaviour. Implementing measurement reveals where the value is and where it isn't.
Strategic priorities. Increase investment in the training that DRIVES the future (data science, AI applications, ESG, customer experience) while reducing investment in training the past (legacy product manuals).
This approach could save £8-10m of the £15m the CEO wants — but without the catastrophic side effects of an indiscriminate cut.
Justified judgement
The CEO's proposal should be rejected as written. Instead, the bank should:
- Conduct a 90-day training audit to identify which training is high-value and which is low-value.
- Cut the genuinely low-value training (probably 30-40% of current spend) — saving £5-7m.
- Reinvest a portion of the saving into higher-value training (data, AI, ESG, customer skills) — net saving £3-5m.
- Implement measurement so future decisions are evidence-based.
This delivers about a third of the CEO's target saving (£3-5m vs £15m) without destroying the firm's capability base.
The likely actual outcome if the CEO insists on the £15m cut:
- Year 1: P&L improves £15m. CEO praised.
- Year 2: Customer complaints rise; minor regulatory issues; first signs of skill gap in newer products. Saving still 'shown'.
- Year 3: Major compliance fine (£50-200m). Loss of senior staff to competitors who invest in their people. Customer experience scores fall.
- Year 4: Recruitment harder. Strategic projects delayed or failing for lack of skills.
- Year 5: The cost of restoring capability is many multiples of the original £15m saving — and competitors who didn't make the cut are now ahead.
The CEO would be remembered as the leader who hollowed out the bank for a one-year P&L gain. That is not the legacy any thoughtful executive wants.
Conclusion. The £15m training cost is real, but cutting it indiscriminately is a textbook short-termist mistake. The right move is a 90-day audit to cut LOW-VALUE training (saving £5-7m) while protecting and even reinvesting in HIGH-VALUE training. The CEO's proposal trades real long-term capability for visible short-term saving — and in a regulated business that competes on expertise, this trade is almost always wealth-destroying.
The deeper insight is that training spend is one of the easiest costs for a new leader to cut visibly, but among the hardest to restore later. The damage is delayed, distributed across many small failures, and only fully visible 3-5 years out — exactly when the cost-cutter has been promoted or moved on, leaving the next leader to clean up the mess. Wise executives resist these 'easy wins' precisely because they are too easy.