On innovation

Abstract

As environments become increasingly uncertain, adaptiveness cannot be sustained through adjustment alone. This essay explores innovation as an organisational capability rather than a discrete activity, examining how innovation depends on the interaction between natural and artificial cognition, and why many organisations struggle to convert intelligence into sustained adaptiveness.

When adaptiveness is no longer enough

Adaptiveness allows organisations to respond to change. It enables adjustment, realignment and course correction as conditions shift. Under moderate uncertainty, this capacity can sustain performance over time.

Under persistent uncertainty, adaptiveness alone becomes insufficient.

As environments continue to evolve, organisations eventually confront conditions for which existing responses no longer apply. Adjusting within current models no longer restores alignment. The organisation must generate new ways of interpreting, deciding and acting.

At this point, innovation becomes necessary not as growth strategy or competitive differentiator, but as a condition of ongoing viability. Uncertainty breeds novel situations and novel situations require an innovative response.

The inherited view of innovation

Most organisations still treat innovation as a bounded activity.

It is associated with functions, programmes or initiatives. Dedicated teams explore new ideas while the core organisation focuses on execution. Innovation is measured through outputs: new products, services, processes or efficiencies.

This approach reflects an implicit assumption: that innovation can be separated from day-to-day operations and added when needed. It presumes that novelty can be produced on demand without altering how the organisation thinks and acts.

This model worked when environments changed slowly and innovation could be scheduled. Novelty could be injected periodically without disrupting the organisation’s underlying logic.

The difficulty arises when the need for innovation is no longer occasional, but continuous.

Innovation as a systemic requirement

Under persistent uncertainty, innovation cannot be confined to isolated units or episodic initiatives. It becomes a systemic requirement.

Innovation, in this context, is not about creativity alone. It is about generating new responses when existing ones no longer suffice. It depends on the organisation’s ability to notice emerging patterns, question assumptions, and recombine insight into novel forms of action.

This capacity is inseparable from how intelligence, adaptiveness and leadership are organised. When sensing is weak, innovation lacks direction. When decision-making is constrained, novelty stalls. When action is tightly controlled, experimentation becomes risky. However, experimentation becomes unavoidable under these conditions.

Innovation is not something added to an organisation. It emerges from how the organisation is designed to think and respond.

Natural cognition as the foundation of innovation

Human cognition remains central to innovation.

People interpret ambiguous signals, imagine alternatives and challenge prevailing assumptions. They connect disparate ideas, draw meaning from experience, and generate insight that cannot be reduced to data alone. Much of this cognition is tacit, contextual and socially mediated.

However, in many organisations, this capacity is underutilised. Structures prioritise execution over enquiry. Time for reflection is scarce. Novel ideas are filtered through layers that favour coherence over exploration.

As a result, natural cognition is often consumed by coordination and compliance rather than sense-making and creation. Innovation is constrained not by a lack of talent, but by how that talent is engaged.

Artificial cognition as an amplifier, not a substitute

Alongside human cognition, artificial cognition now plays an increasingly significant role.

Data analytics, machine learning and AI systems can detect patterns, process scale and surface correlations beyond human capacity. They extend the organisation’s sensing and analytical reach.

Yet artificial cognition does not innovate independently. It amplifies the intelligence model into which it is embedded.

When organisations treat innovation as a technical problem, AI is positioned as a substitute for human insight. When they treat intelligence as organisational, AI becomes an amplifier of collective sense-making.

Artificial cognition can expand what the organisation notices and explores. It cannot determine what matters or why.

Innovation emerges not from replacing human cognition, but from the interaction between natural and artificial forms of intelligence within an enabling system.

Why innovation efforts often stall

Despite significant investment, many organisations struggle to sustain innovation.

Common explanations focus on culture, incentives or risk appetite. While these factors matter, they often mask a deeper issue: innovation is being asked to operate within organisational designs optimised for stability and efficiency.

Ideas are generated but cannot be acted upon. Experiments are launched but constrained by approval processes. Insight accumulates but fails to translate into change.

Innovation becomes performative rather than productive.

The problem is not resistance to change. It is a system that constrains how new understanding is explored, tested and integrated. Innovation efforts falter not because people lack imagination, but because the organisation lacks the capacity to absorb novelty.

Two different paths

Embodied innovation

At Amazon, an e-commerce, cloud computing and digital services business, innovation is not treated as a discrete activity or protected function. It is designed as a system-level capability, embedded in how teams sense demand, make decisions, and act under uncertainty. Small, autonomous teams are given clear domains, decision authority, and direct exposure to customer signals. Failure is expected and structurally absorbed through modular design, fast feedback loops, and decoupled architectures. Innovation does not rely on prediction or executive sponsorship; it emerges because the organisation is built to explore, test, and discard ideas continuously without destabilising the whole. Innovation capacity scales because it is distributed rather than centralised.

Innovation constrained

At Xerox, at its peak an office photocopying and document technology, as well as a tech research giant, innovation was organisationally separated from execution. Breakthrough ideas were generated within dedicated research units, but judgement and authority remained tied to the core business model. Signals about future markets existed, but decisions about commercialisation were evaluated through legacy assumptions about customers, revenue, and risk. Innovation was treated as something to be handed over, rather than something the organisation itself could enact. As a result, intelligence was generated but not converted into adaptive action. Innovation existed, but it could not survive contact with the organisation’s operating logic.

The contrast

The difference between Amazon and Xerox was not creativity, technical brilliance, or access to ideas. It was how innovation was structurally positioned. At Amazon, innovation was integrated into the organisation’s decision and action loops, allowing exploration without threatening coherence. At Xerox, innovation was isolated from those loops, leaving judgement anchored in the past. When uncertainty increased, Amazon’s innovation capacity compounded. Xerox’s dissipated. Innovation proved to be less about invention than about whether organisational design allows new intelligence to be acted upon.

Innovation as sustained adaptive capacity

If adaptiveness is to be sustained, innovation must be understood differently.

It is not a pipeline that feeds execution, nor a lab that operates at the margins. It is an ongoing capacity to generate new responses as conditions evolve. This capacity depends on how sensing, intelligence, leadership and adaptiveness interact across the organisation.

Natural and artificial cognition together shape what the organisation can perceive and imagine. Whether that imagination becomes action depends on organisational design.

What it would mean to design organisations where innovation is a continuous property rather than a discrete function remains unresolved. What is increasingly evident is that sustained adaptiveness depends on innovation that is woven into the organisation itself.