Analyses, observations and practical frameworks on ownership, capital, governance, organization and transformation.
Themes
Management, delegation, founder dependency and leadership teams.
Related perspectives
Content related to this theme.
Open Finance and agentic AI could progressively unbundle what banks currently provide as one integrated proposition. As information and analysis become machine-accessible, value may concentrate at two ends: hard financial infrastructure, and the harder-to-commoditise work of understanding objectives, making trade-offs and assuming responsibility.
Private banks have become highly sophisticated at differentiating client needs, yet RM allocation still relies heavily on market, availability, relationship history and past performance. A missing variable is whether the way an RM creates value actually matches what a particular client needs.
Hunter/Farmer answers an important question about where growth comes from, but it captures only one dimension of RM value creation. Expertise, judgement, trust, influence and orchestration can differentiate RMs even when their acquisition/development profile looks similar.
Two RMs can produce the same numbers through very different forms of value creation. That distinction matters when past performance is used to decide whom to hire, which clients to allocate, what to develop or how to prepare succession.
AI can compress organizational loops from sensing to learning, but speed creates a new risk: good outcomes can be mistaken for proof that the reasoning behind them was correct. AI-native organizations therefore need an explicit architecture for distinguishing what is known, assumed, tested and actually learned.
Private banks measure RMs through book size, revenues, NNM and activity, but those metrics say little about why clients value one RM over another. Making different modes of relationship value visible can improve client allocation, talent development and the bank's ability to retain what its people know.
Once the outcome is known, hindsight rewrites assumptions and makes uncertainty disappear. Recording expectations, timing, critical conditions and what would change your mind before a decision creates a basis for comparing reasoning with reality rather than reconstructing the story afterwards.
Every important decision contains a hypothesis about what will happen next. Making assumptions, causal mechanisms and conditions explicit improves the decision before it is made and makes learning possible when reality responds.
People infer strategy from what gets funded, rewarded, escalated, tolerated and repeatedly questioned. Organizational reasoning makes those signals explicit and compares declared intent, observed behaviour and actual outcomes to see whether they tell the same story.
As AI tools and models become more accessible, advantage shifts from technology to organizational capability: identifying where AI changes business economics, redesigning work and decisions, governing those systems and learning from outcomes. The deeper layer is organizational reasoning.
Employees do not learn strategy only from formal communication. They infer it from the repeated signals leaders send through what they value, question, fund, escalate, tolerate, reward and follow through. Making that implicit logic visible is one of the foundations of organizational reasoning.
KPMG’s latest Global AI Pulse shows AI adoption accelerating faster than value realization. As model capability improves, the constraint increasingly becomes whether organizations have made their decision logic explicit enough for humans and machines to reason within it together.
Organizations shape behavior through what gets rewarded, punished, promoted, ignored and escalated. Those feedback loops reveal the operating logic people actually learn.
AI will scale the decision logic an organization already has. If that logic remains implicit, inconsistent or contradictory, automation risks turning local weaknesses into systemic ones.
ITRI’s role in Taiwan’s semiconductor ecosystem suggests that durable competitive advantage may lie less in individual winners than in institutions that continuously build capabilities, transfer knowledge and improve collective reasoning.
Strategy, governance, processes and AI all depend on an underlying decision logic. Making that organizational reasoning visible, testable and improvable could become a defining capability of AI-native organizations.
A good decision creates value once. A decision system makes better judgment, faster decisions and more consistent execution repeatable across the organization.
Private banks often define service levels clearly but rarely measure the effort consumed by each relationship. Without that visibility, high-touch can mean premium service - or an unpriced subsidy.
Private-bank margin erosion is often a complexity problem rather than a single cost problem. Measuring effort by client reveals what to price, standardize, automate, migrate - or stop.
Complexity rarely appears as one visible cost. It accumulates through exceptions, bespoke service, controls and senior attention, quietly consuming margin.
Profit is not produced evenly across a business. Understanding where economic value is created — and where it is absorbed — is essential to managing performance.
Better decisions become possible when reality becomes visible. What an organization can see clearly shapes what it can decide, prioritize and change.
Private banks know their AuM and headline productivity metrics, but averages often conceal where revenue is produced, effort is consumed, complexity is absorbed and future client value is being created or neglected.
Smaller Swiss private banks appear to have converted AuM into profit more effectively than larger institutions. The lesson is not that small is better, but that focus, client mix, pricing, transformation costs and complexity shape the economics of scale.