The New Race for Organisational Intelligence: Implications for Managers

By Graeme Leversha and J. Mark Munoz

Artificial intelligence alone will not determine future competitive advantage; organisations that learn to understand themselves may become tomorrow’s market leaders.

Organisations are investing unprecedented sums in artificial intelligence, yet many still struggle to understand themselves. This article argues that the next competitive race is not simply for better AI, but for greater Organisational Intelligence – the ability to understand an organisation and its environment in pursuit of its purpose. Graeme Leversha of Frontier Partners Ltd and J. Mark Munoz of Millikin University explore why this shift matters for managers.

Optimising Organisational Intelligence

Organisational Intelligence is not a new term. Its intellectual history extends for more than half a century and draws upon several important traditions in management and organisational theory. Harold Wilensky’s Organizational Intelligence: Knowledge and Policy in Government and Industry (1967) examined how organisations acquire, process, and use information in decision-making. Russell Ackoff established the foundations of purposeful systems thinking. Chris Argyris and Donald Schön examined how organisations learn. David Teece and colleagues developed the concept of dynamic capabilities to explain how organisations adapt to changing environments. Thomas Malone and others have explored collective intelligence and the capacity of groups to solve problems together.

Each of these traditions has contributed important insights into aspects of organisational intelligence. They have largely been developed independently. What has remained less clear is how these capabilities relate to one another and whether they can be brought together within a coherent theory of the intelligent organisation.

Recent research undertaken through the Horizon Organisational Intelligence Research Programme¹ proposes that Organisational Intelligence can be understood through the interaction of three foundational capabilities: Purpose, Representation, and Understanding. This framework draws on and synthesises the established traditions above.

OI = f(P, R, U)

Figure 1: The Organisational Intelligence Model.

Organisational Intelligence emerges from the interaction of Purpose, Representation, and Understanding. Better understanding supports better judgement, decisions, outcomes, and learning, which in turn improves the organisation’s representation of itself.

Purpose establishes why the organisation exists and what it seeks to achieve. Representation provides a coherent model of what the organisation is – its identity, capabilities, structures, relationships, authorities, and obligations. Understanding enables the organisation to interpret that representation within the context of its environment and experience.

None is sufficient alone. Purpose without representation has no coherent organisational object through which it can be pursued. Representation without understanding creates description rather than intelligence. Understanding without purpose provides no basis for determining what matters.

It is through their interaction that the capabilities normally associated with intelligent organisations begin to emerge. An organisation can discover changes in itself and its environment, exercise judgement about their significance, learn from experience, adapt its behaviour, and govern its actions.

This leads to a practical definition:

An intelligent organisation is one capable of maintaining and improving its understanding of itself and its environment in pursuit of its purpose.

The importance of this definition lies not simply in what an intelligent organisation knows, but in what happens next. Organisational understanding has the potential to become cumulative. Better representation enables better understanding. Better understanding supports better judgement. Better judgement informs action. Action creates experience. That experience can then improve the organisation’s representation of itself, creating the conditions for deeper understanding.

Representation → Understanding → Judgement → Action → Experience → Improved Representation

This feedback loop matters because it suggests that Organisational Intelligence may compound over time, creating a potentially durable source of competitive advantage.

For managers, this reframes the AI opportunity. Organisational Intelligence is not simply about deploying more AI. It is about improving the organisation’s capacity to understand what it is, what is happening to it, why it is happening, and what it should do next.

Assessing Organisational Intelligence

Organisations need to understand their organisational intelligence capability. It is important to evaluate how close they are to the intelligent organisation described above. Despite significant investment in enterprise technology and data infrastructure, the gap between information availability and organisational understanding remains significant. Research by McKinsey found that the majority of large organisations report difficulty connecting strategic intent to operational behaviour, with fewer than 30 percent of transformation programmes delivering sustained value.

Drawing directly from the Purpose–Representation–Understanding (PRU) framework developed above, we propose a simple managerial diagnostic: the OI-PRU Assessment. It translates the three foundational capabilities of Organisational Intelligence into three questions that organisations can use to identify where their most significant intelligence gaps may lie.

Most organisations will answer “no” to at least one of these questions – and many will answer “no” to all three. The gap has historically been treated as a management challenge. It is increasingly becoming a competitive and governance challenge as well.

Why Don’t Organisations Understand Themselves Better?

There is a curious variance at the centre of modern enterprise technology. Organisations have become exceptionally good at representing the things they manage. For instance, customer relationship systems represent customers. Enterprise resource planning systems represent transactions and resources. Human resource systems represent employees. Governance, risk, and compliance (GRC) platforms represent risks and controls. Regulatory systems represent obligations.

Yet, the organisation itself typically remains implicit. Its purpose may exist in a strategy document. Its policies may sit in a document management system. Its authorities may be maintained in delegations. Its risks may be held in a GRC system. Its institutional memory may be dispersed among documents, emails, systems, and people. Organisational information exists everywhere but it is not sensibly and systematically organised.

This scenario creates a fundamental problem for the application of artificial intelligence. An AI system may possess extraordinary reasoning capabilities, but if it lacks a coherent representation of the organisation in which it operates, it cannot fully understand the organisational significance of its actions. As AI deployments scale, this gap becomes increasingly consequential. Research from Gartner suggested that a majority of AI projects fail to deliver on their expected value, with a lack of organisational context frequently identified as a contributing factor, rather than model capability alone.

An AI agent operating inside a bank, for example, may be technically capable of approving a transaction, extending credit, or moving capital. But capability does not establish authority. The organisational questions are different. Is the action consistent with the bank’s purpose and policy? Does the agent have delegated authority? Have regulatory obligations been satisfied? How does the decision affect the bank’s risk position? Is human approval required? A machine may understand the task without understanding the organisation.

This leads to the concept of Computational Organisational Representation, or COR. COR proposes that the organisation itself can be represented as a persistent computational entity, encompassing its identity, purpose, capabilities, governance, relationships, authorities, obligations, and institutional knowledge as a coherent whole.

The distinction from a digital twin is important. A digital twin typically represents the changing state of an entity or system. A COR establishes the persistent entity whose state is changing. The COR answers: what is this organisation? A digital twin then helps answer: what condition is it currently in? Together, they create the possibility of a continuously evolving understanding of the organisation itself.

Can Organisational Intelligence Become Computational?

If Organisational Intelligence describes the capability and COR provides the representational foundation, the next question is whether these concepts can be realised computationally.

An Intelligent Organisational System, or IOS, is not an AI system that happens to operate within an organisation. It is a system that operates with explicit organisational context, reasoning with reference to purpose, authority, policy, history, and current organisational state. The progression from theory to engineering is:

Organisational Intelligence → Computational Organisational Representation → Intelligent Organisational Systems

This progression matters because increasingly capable AI does not reduce the need for organisational context. It makes that context more important. As AI systems move from generating information to recommending decisions and eventually taking autonomous action, organisations will need computational mechanisms capable of determining not simply what an AI system can do, but what it should be permitted to do.

One approach is being explored through Horizon, a reference architecture being developed by Frontier Partners to investigate the computational realisation of Organisational Intelligence. Horizon investigates how persistent organisational representation, institutional memory, contextual reasoning, runtime governance, assurance, and controlled organisational action might operate together within a computational architecture. The theory of Organisational Intelligence does not depend upon Horizon or upon any commercial implementation. Rather, Horizon represents one attempt to test whether aspects of Organisational Intelligence can be realised and evaluated computationally. If Organisational Intelligence develops as a recognised discipline, multiple architectures and implementations should ultimately emerge.²

Navigating the New Age of Organisational Intelligence

There may be a broader historical context for this development. Russell Ackoff described the transition from the Machine Age to the Systems Age as a fundamental change in how complex phenomena were understood. The Machine Age, shaped by the Industrial Revolution, emphasised reductionism and mechanism. The Systems Age introduced a different perspective: the behaviour of the whole could not be understood by studying its components independently. Relationships, interactions, purpose and the larger systems within which entities existed became central.

Organisational Intelligence may represent a further development in this progression.

The Machine Age taught organisations to optimise their parts. The Systems Age taught organisations to understand the relationships between those parts. The emerging Organisational Intelligence Age may enable organisations to understand themselves.

This does not imply organisational consciousness or sentience. It underscores the question of whether an organisation can maintain a sufficiently coherent understanding of its purpose, structure, current condition, experience, and environment to improve its judgement and behaviour over time. That is a proposition that can increasingly be tested.

The Compounding Advantage of Organisational Intelligence

The commercial implications of Organisational Intelligence may be significant, particularly where organisational understanding becomes a source of competitive advantage. Consider two banks operating in the same market. Both have access to similar AI models, cloud infrastructure, and data analytics. Both can purchase comparable technologies and recruit capable people. As advanced AI becomes widely available, access to the model itself may provide only a temporary competitive advantage.

The more enduring advantage may belong to the bank that understands itself better. Such a bank may identify emerging risks earlier, recognise deteriorating controls before they produce material losses, understand why customer outcomes are changing, preserve institutional knowledge when employees leave, and learn systematically from previous decisions.

An organisation that learns faster than its competitors may become progressively better at learning.

The compounding mechanism described earlier now becomes commercially significant. Better representation enables better understanding. Better understanding produces better judgement and better outcomes. Those outcomes create experience that can improve the organisation’s representation of itself, enabling still deeper understanding. The organisation does not simply accumulate more information; it becomes progressively better at understanding what that information means. This mirrors dynamics well understood in financial markets, where even small analytical advantages compound significantly when they consistently produce better decisions over time.³

An organisation that learns faster than its competitors may become progressively better at learning. An organisation that understands risk better may become progressively better at taking risk. An organisation that understands its customers better may become progressively better at serving them.

The contemporary competitive race may therefore increasingly become a race in Organisational Intelligence.

Implications for Managers

Organisational Intelligence changes the managerial question from how much AI should we deploy? to how can AI improve the organisation’s capacity to understand and improve itself? This has four immediate implications.

• Develop a holistic strategy. AI strategy increasingly becomes organisational strategy. As access to powerful AI models becomes widespread, competitive advantage is more likely to arise from the organisational context in which those models operate than from the models themselves. Managers need to work from a high-level perspective and integrate their AI objectives with their overall vision, mission, and goals.
• Plan for an optimal organisational architecture. Organisational representation becomes strategic infrastructure. Managers should aim to create an organisational structure that can coherently represent its purpose, capabilities, authority, obligations, risks, and institutional knowledge, rather than leaving departments or divisions fragmented across systems and people.
• Make governance a priority. Greater AI autonomy requires stronger organisational context and governance. An AI system may be capable of taking an action without understanding whether it has the authority to do so, whether the action is consistent with organisational purpose, or how it changes organisational risk. Managers need to be cognizant of this context and prioritise a plan for impactful management practice and governance.
• Leverage organisational intelligence. Organisational understanding can become a potential source of compounding competitive advantage. Managers that represent themselves more accurately can understand themselves better, make better decisions, and learn more effectively from the resulting experience.

The key managerial goal therefore becomes increasingly clear: build an organisation capable of understanding and improving itself faster than its competitors.

Concluding Thoughts

Scientific discoveries, technological innovations, and economic advances may originate with individuals, but they rarely create enduring value through individuals alone. They are developed through universities, laboratories, companies, governments, and research institutions. Knowledge becomes economically and socially consequential when organisations preserve it, combine it with existing knowledge, govern its application, and learn from its consequences.

Organisations are among humanity’s most important mechanisms for accumulating intelligence across time.

This suggests that the current race towards increasingly capable artificial intelligence may be only one part of a larger transition. Artificial intelligence asks how machines can become more intelligent. Collective intelligence asks how groups can become more intelligent. Organisational Intelligence asks how organisations themselves can develop greater understanding, and whether that understanding can compound in ways that create durable advantage.⁴

As artificial intelligence becomes increasingly accessible, the decisive competitive question may no longer be which organisation possesses the most powerful model. Similar models may eventually be available to everyone.

The more enduring advantage may belong to the organisation that can represent itself more accurately, understand itself more deeply, learn faster from experience, and adapt more intelligently than its competitors.

The race for artificial intelligence will continue.
But the larger prize may ultimately be
Organisational Intelligence.

About the Authors

Graeme Leversha is the founder and Managing Director of Frontier Partners Limited, where he leads the development of Horizon and the Horizon Organisational Intelligence Research Programme. His work focuses on Organisational Intelligence, AI governance, and the computational architecture of intelligent organisations.

J. Mark Munoz is Professor of Management at Millikin University and a former Visiting Fellow at Harvard University. A recipient of several awards, including the ACBSP Teaching Excellence Award and the AGBA Lifetime Achievement Award in Global Business, he has authored and edited more than 30 books, including The AI Leader, Global Business Intelligence, and the Handbook of Artificial Intelligence and Robotic Process Automation. His research interests include artificial intelligence, management, innovation, and emerging technologies. He serves on the Advisory Board of Frontier Partners Limited.

References

1. Leversha, G. (2026). “Organisational Intelligence: A Foundational Theory”. Horizon Organisational Intelligence Research Programme Working Paper. Frontier Partners Limited, Auckland.

2. Horizon Organisational Intelligence Research Programme (2026). “Reference Architecture Overview”. Frontier Partners Limited. Internal Research Document.

3. Teece, D.J., Pisano, G. and Shuen, A. (1997). “Dynamic capabilities and strategic management”. Strategic Management Journal, 18(7), 509–33.

4. Malone, T.W. (2018). Superminds: The Surprising Power of People and Computers Thinking Together. New York: Little, Brown and Company.