Editor’s note: This is the 21^st article since May 20, 2026 in an ongoing series by Dr. Andrew Maxwell, the Bergeron Chair in Technology Entrepreneurship in the Lassonde School of Engineering at York University. Every week – and occasionally every other week – we’ll present a new article by Maxwell, in a series whose wide-ranging and incisive themes encompass: Canada and innovation policy; productivity and industry; innovation frameworks; AI and higher education; research and intellectual property; technology adoption; entrepreneurship and commercialization; universities and higher education; entrepreneurship education; and AI and the future of work. 

This is Part 4 of a four-part series on innovation, leadership and organizational adaptation. Part 1 was published on September 30, 2026, Part 2 on October 7, and Part 3 on October 14.

For decades, good governance meant minimizing risk, benchmarking against peers, and avoiding visible failure.

That logic worked in a relatively stable world.

But in environments shaped by accelerating technologies, geopolitical uncertainty, and AI-enabled disruption, the greatest risk may now be the risk of doing nothing.

This article concludes a short series exploring why many capable organizations struggle to adapt – even when they recognize that change is necessary.

Across the series, a troubling pattern emerged.

Many organizations are not failing to innovate because they lack ideas, talent or resources.

They struggle because their institutions were designed for stability, not adaptation.

The arc of the argument

The previous articles explored different aspects of this paradox.

The first article asked a simple but uncomfortable question: are many organizations structurally designed not to innovate?

In pursuing reliability, predictability and defensibility, organizations often develop governance structures that unintentionally suppress experimentation.

The second piece examined how we measure innovation.
Traditional metrics – patents, R&D spending, benchmarking against peers – can reinforce incremental thinking and discourage the kind of experimentation required in uncertain environments.

The third article explored a deeper tension: when legitimacy slows learning.

Institutions that must maintain trust, credibility, and accountability often face the strongest pressures to avoid visible failure. The result is a system where experimentation becomes difficult precisely where adaptation is most needed.

This final piece brings these threads together and asks the central question:

How do we design institutions that can maintain legitimacy while learning fast enough to adapt to volatility?

The real constraint: Learning velocity

In stable environments, the objective of institutions is efficiency.

Processes are optimized. Variability is reduced. Risk is minimized.

But volatility changes the nature of the challenge.

When technologies evolve rapidly and uncertainty is high, the central constraint shifts.

The most important capability becomes learning velocity – the rate at which an organization can reduce uncertainty.

Organizations that learn faster adapt faster.

Those that learn slowly fall behind.

Yet many institutional structures unintentionally slow learning.

Approval processes delay experimentation.
Risk evaluation frameworks penalize failure.
Benchmarking encourages imitation rather than exploration.

Even well-intentioned governance systems can become bottlenecks.

The behavioral dimension of institutional inertia

Part of this challenge is structural, but part is deeply human.

Research in behavioural economics shows that decision-making under uncertainty is strongly shaped by loss aversion and status quo bias.

Potential losses loom larger than equivalent gains.

In organizational contexts, this bias is amplified by institutional pressures.

Leaders must defend decisions.
Boards expect predictability.
Public organizations must justify expenditures.

In this environment, avoiding visible failure often becomes more important than reducing long-term uncertainty.

The result is a subtle but powerful bias toward delay.

The paradox is that while experimentation carries visible short-term risk, inaction carries cumulative long-term risk.

But this risk is often harder to see.

The risk of doing nothing

This insight echoes an idea from Israeli business management expert Eliyahu Goldratt’s Theory of Constraints, which highlights how systems often focus on optimizing local efficiency rather than addressing the true bottleneck.

In many contemporary institutions, the bottleneck is not capability.

It is the speed of learning.

When experimentation slows, uncertainty persists.
When uncertainty persists, decision-making becomes cautious.
When caution dominates, adaptation stalls.

The real risk is not experimentation.

The real risk is institutional delay.

Recognizing this challenge does not mean abandoning stability.

Institutions exist for good reasons.

Governance structures protect trust.
Standards ensure reliability.
Accountability preserves legitimacy.

The challenge is not to replace stability with chaos.

The challenge is to separate stability from exploration.

Adaptive institutions recognize that these two logics must coexist.

Core operations require reliability and discipline.

Exploratory activities require flexibility and experimentation.

The problem arises when a single governance logic governs both.

When systems designed to ensure stability are applied to exploratory learning, experimentation becomes almost impossible.

Designing adaptive institutions

If the challenge is architectural rather than cultural, what might adaptive institutions look like?

Three design principles appear particularly important.

First, differentiate stability and exploration.

Core operations should maintain rigorous governance structures that protect reliability and accountability. Experimental initiatives should operate under lighter frameworks designed to encourage learning and iteration.

Second, evaluate experiments differently from operations.

Operational activities are judged on performance outcomes. Experiments should be judged on learning outcomes – whether uncertainty has been reduced and insights generated.

Third, accelerate learning cycles.

Technologies such as artificial intelligence now offer powerful opportunities to simulate, test, and iterate ideas more rapidly than traditional approaches allow.

Used thoughtfully, these tools can significantly increase learning velocity.

But technology alone is not sufficient.

Without institutional redesign, AI risks becoming simply another optimization tool rather than a catalyst for exploration.

From innovation programs to institutional design

For many organizations, the response to innovation challenges has been to create new programs.

Innovation labs. Accelerators. Corporate venture funds.

These initiatives can be valuable, but they rarely address the deeper issue.

Innovation problems are often not program problems.

They are institutional design problems.

If governance structures discourage experimentation, even well-funded initiatives struggle to deliver meaningful impact.

What is required instead is a shift in perspective.

Innovation is not simply about generating ideas.

It is about designing institutions that can learn continuously.

These reflections were originally developed for the International Society for Professional Innovation Management community, which for decades has brought together scholars and practitioners to explore the evolving nature of innovation systems.

Today that conversation feels more urgent than ever.

The challenges we face – technological disruption, climate transition, geopolitical change – require institutions capable of adapting quickly while maintaining legitimacy and trust.

Designing such institutions will not be easy.

But it may be one of the most important innovation challenges of our time.

References

Cyert, R. M., & March, J. G. (1963). A behavioral theory of the firm. Prentice-Hall.

Goldratt, E. M. (1990). The Haystack Syndrome. North River Press.

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291.

March, J. G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2(1), 71–87.

North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge University Press.

Teece, D. J. (2018). Business models and dynamic capabilities. Long Range Planning, 51(1), 40–49.

R$