Enabling evolutionary innovation

The ability to shape conditions that allow useful novelty to emerge through recombination, exaptation, and exploration of the adjacent possible, rather than attempting to design, predict, or plan innovation in advance.

Leaders skilled in enabling evolutionary innovation recognise that innovation in complex adaptive systems is not an act of invention, but a process of emergence. Novelty arises through variation, interaction, and selection over time, not through linear roadmaps or speculative forecasting. The leadership task is therefore not to define the breakthrough, but to create the conditions in which new uses, combinations, and directions can be discovered.

Instead of pursuing radical ideas in isolation, these leaders focus on making the next viable step possible. They work with what already exists, current capabilities, data, relationships, and practices, and explore how these might be repurposed or recombined in response to shifting conditions. Innovation is treated as evolutionary movement, not a leap into the unknown.

This capability represents a shift from directing innovation to hosting it. Leaders resist the urge to demand certainty, business cases, or long-term plans for ideas that have not yet had a chance to form. They design environments that encourage low-cost experimentation, cross-boundary interaction, and rapid learning through use, allowing novelty to emerge where information is richest.

At its core, enabling evolutionary innovation expands the system’s capacity to discover what works next. Leaders do this by increasing variation, enabling connection, protecting early signals from premature judgement, and allowing selection to occur through real-world feedback. Progress is achieved not through dramatic invention, but through small, intelligent steps that accumulate into sustained adaptation over time.

“Chance favours the connected mind.” – Steven Johnson

Why enabling evolutionary innovation matters

Enabling evolutionary innovation matters because, in complex adaptive systems, the future cannot be reliably predicted or designed in advance. When leaders attempt to plan innovation through long-range roadmaps, detailed business cases, or singular breakthrough bets, they force the system to commit to assumptions that have not yet been tested. This increases risk while reducing adaptability.

From a leadership perspective, evolutionary approaches reduce exposure while increasing learning. Small, adjacent steps allow ideas to be explored without requiring early certainty or large commitments. Rather than betting heavily on speculative futures, leaders enable multiple low-cost experiments and allow evidence to emerge through use. This shifts innovation from a high-stakes gamble to a managed process of discovery.

Enabling evolutionary innovation also increases speed without sacrificing judgement. Linear innovation models slow decision-making by demanding justification before insight exists. Evolutionary approaches move faster because they allow action to precede explanation. Leaders gain momentum by letting the system try, sense, and adjust, rather than waiting for confidence that never arrives.

This capability matters because it preserves coherence under uncertainty. When leaders chase grand visions or impose top-down innovation agendas, effort fragments and credibility erodes as reality diverges from plan. By contrast, evolutionary innovation keeps activity grounded in existing capabilities and real feedback, allowing change to emerge without destabilising the system.

Finally, enabling evolutionary innovation strengthens long-term system viability. Organisations that rely on episodic breakthroughs become brittle when conditions shift. Those that continuously explore the adjacent possible build a steady capacity to adapt. Leaders who enable this capability ensure the system is not dependent on rare moments of inspiration, but is always learning its way forward.

“Novel technologies arise by combination of existing technologies.” Brian Arthur

What good and bad looks like for enabling evolutionary innovation

Weak enabling of evolutionary innovation (engineered certainty or suppression)	Strong enabling of evolutionary innovation (designed evolutionary conditions)
Seeks breakthrough ideas: Looks for big, original concepts detached from current reality.	Creates conditions for emergence: Shapes environments where useful novelty can arise through interaction and use.
Demands clarity before movement: Expects ideas to be fully formed and justified before action.	Allows movement to create clarity: Uses action and experimentation to surface insight that cannot be predicted in advance.
Treats innovation as an initiative: Frames innovation as a bounded programme with plans and milestones.	Treats innovation as continuous: Accepts innovation as an ongoing process of variation and selection.
Future fixation: Pushes attention towards distant, speculative futures.	Adjacent possible focus: Directs attention to the next viable step that current conditions make possible.
Prediction-led evaluation: Judges ideas through forecasts, business cases, or assumed outcomes.	Feedback-led evaluation: Judges ideas through real-world use, signals, and learning.
Centralises innovation direction: Retains control over what should be explored and why.	Enables local discovery: Allows exploration where information and constraints are richest.
Interrupts ambiguity: Shuts down messy or unclear work to restore order or confidence.	Protects early exploration: Shields immature ideas from premature judgement.
Optimises relentlessly for efficiency: Eliminates slack, redundancy, and unused capacity.	Preserves slack for learning: Maintains spare capacity to allow tinkering and recombination.
Suppresses deviation: Treats departures from plan as errors to correct.	Harvests deviation: Treats anomalies and workarounds as potential signals of new direction.
Measures success by plan delivery: Values execution against predefined expectations.	Measures success by learning rate: Values speed of insight and options created.

“When you step into an intersection of fields, disciplines, or cultures, you can combine existing concepts into a large number of extraordinary new ideas.” – Frans Johansson

Barriers to enabling evolutionary innovation

Premature convergence: Leaders push too quickly toward a single “good idea” or preferred direction. This collapses variation before the system has had a chance to explore multiple paths, undermining evolutionary discovery.

Originality bias: Leaders privilege novel, impressive ideas over recombination and reuse. By equating innovation with originality, they overlook the evolutionary power of exaptation and adjacent possibilities.

Narrative lock-in: Once a compelling innovation story takes hold, leaders defend it even as evidence shifts. This reduces selection pressure and prevents the system from abandoning paths that are no longer viable.

Early filtering of weak signals: Small, incomplete, or awkward ideas are dismissed because they lack clarity or polish. Evolutionary innovation depends on weak signals being allowed to develop before their value is obvious.

Over-designing the future: Leaders attempt to specify distant outcomes rather than enabling near-term exploration. This pulls attention away from what is possible now and increases commitment to untestable assumptions.

Suppression of deviation: Workarounds, hacks, and unofficial solutions are treated as non-compliance rather than potential sources of innovation. Evolutionary pathways are closed rather than examined.

Expert gravity: Established expertise shapes what is seen as plausible. Leaders defer to dominant paradigms, reducing variation and blinding the system to possibilities that sit outside current frames.

Innovation exceptionalism: Innovation is treated as a special activity requiring permission, funding rounds, or designated forums. This slows exploration and prevents everyday experimentation from occurring where insight is richest.

Fear of visible inefficiency: Evolutionary processes appear messy and wasteful. Leaders concerned with optics or reputational exposure shut down exploration that does not immediately look productive.

Selection without learning: Leaders make keep-or-kill decisions without understanding why an idea failed or succeeded. This stops the system from accumulating evolutionary knowledge and repeating useful patterns.

“Innovation comes from trial and error, tinkering, and experimentation.” – Nassim Taleb

Enablers of enabling evolutionary innovation

Protection of variation before convergence: Leaders deliberately prevent early narrowing of options by resisting pressure to select a single “best idea” too soon. They use their authority to keep multiple possibilities alive long enough for learning to occur, especially when ambiguity is high.

Recombination mindset: Leaders consistently direct attention to how existing assets, capabilities, data, or relationships might be reused or combined in new ways. Innovation is framed as recombination, not invention, reducing dependence on originality and lowering the cost of exploration.

Legitimisation of weak signals: Leaders explicitly protect early, incomplete, or awkward ideas from premature dismissal. They signal that partial insights are valuable inputs to exploration, even when they cannot yet be justified or explained.

Narrative looseness: Leaders hold innovation narratives lightly and resist locking the organisation into a single story about the future. They allow meaning and direction to evolve as evidence accumulates, rather than defending a compelling but fragile vision.

Attention to deviations and workarounds: Leaders pay deliberate attention to informal practices, hacks, and local adaptations. Instead of correcting them, they ask what problem is being solved and whether the deviation reveals a broader opportunity.

Permissionless local experimentation: Leaders make it clear that small, low-risk experiments do not require special approval or formal innovation processes. This keeps exploration close to the work and prevents innovation from becoming centralised or ceremonial.

Delayed evaluation discipline: Leaders consciously separate exploration from judgement. They delay keep-or-kill decisions until sufficient feedback exists, preventing prediction-based selection from replacing learning-based selection.

Selective amplification: Leaders watch for emerging patterns that show promise and amplify them through attention, resources, and connection. Direction emerges through reinforcement, not instruction.

Tolerance for visible inefficiency: Leaders accept that evolutionary innovation appears inefficient in the short term. They protect exploratory activity from optimisation pressure long enough for value to become visible.

Retention of learning, not just outcomes: Leaders ensure that insights from experiments, successful or not, are retained and reused. What matters is not only which ideas survive, but what the system learns about how novelty emerges and evolves.

“Breakthrough innovations often come from outside the company, or from unexpected places inside it.” – Clayton Christensen

The 10 best self-reflection questions for enabling evolutionary innovation

When new ideas emerge, do I instinctively narrow toward a single “best” option, or do I deliberately keep multiple possibilities alive long enough for learning to occur?

How often do I require clarity, justification, or a business case before allowing exploration to begin?

Do I primarily look for impressive, original ideas, or do I actively search for ways existing assets, practices, or data could be repurposed or recombined?

Once a compelling innovation story forms, how willing am I to let it change as evidence accumulates?

What weak, incomplete, or awkward ideas have I dismissed recently because they did not yet make sense?

Do I treat workarounds, hacks, and unofficial solutions as problems to eliminate, or as signals worth examining?

How easy is it in my environment for someone to try a small experiment without seeking permission, sponsorship, or formal approval?

When ideas are evaluated, does judgement usually occur before real-world use, or after feedback has been generated?

How tolerant am I of visible inefficiency, duplication, or exploration that does not immediately look productive?

When I amplify an idea, am I responding to evidence emerging from use, or to alignment with my preferred direction?

“How can I know what I think until I see what I say?” – Karl Weick

Micro-practices for enabling evolutionary innovation

1. Delay selection by one meeting

In complex systems, the greatest threat to innovation is not lack of ideas but premature selection. Senior leaders are powerful selectors. The moment a leader endorses, funds, or favours an option, variation collapses and the system commits before it has learned enough. Practise deliberately delaying selection by one decision cycle. When presented with multiple options, explicitly say: “We are not choosing today. I want one more round of exploration, use, or learning before we converge.”

This does not mean inaction. Teams can still test, prototype, or probe. What is delayed is commitment, not movement. This practice keeps evolutionary space open just long enough for weak signals to strengthen, evidence to surface, and better options to emerge. It reduces later reversals, protects credibility, and increases the quality of strategic commitment without slowing momentum.

2. Replace evaluation with curiosity

Senior leaders often kill evolutionary potential unintentionally through evaluative reflexes. Comments like “That won’t scale” or “We tried that before” feel efficient but act as early selection pressure, stopping ideas before they can evolve. Practise replacing evaluation with curiosity. When an idea is presented, resist judging its merit and instead ask:

• “What would we need to learn to know whether this is interesting?”
• “What would have to be true for this to be useful?”
• “What is this a first step towards?”

This shifts the conversation from justification to exploration. It allows ideas to mutate, combine, or redirect rather than being accepted or rejected as-is. Over time, this practice changes what people bring forward. Instead of polished, defensive proposals, you receive early-stage thinking, hunches, and partial insights, exactly the raw material evolutionary innovation depends on.

3. Protect one weak signal

Most breakthrough innovations begin as weak, awkward, or incomplete ideas. They lack language, data, or status and are therefore easy to dismiss. Leaders can dramatically alter innovation outcomes by protecting just one weak signal at a time. When an idea feels underdeveloped or unpopular but intriguing, say explicitly: “This isn’t clear yet, but I don’t want us to lose it. Let’s keep this alive.” Do not force it to justify itself. Do not demand a business case. Simply ensure it is revisited.

This practice changes selection dynamics. It signals that novelty does not have to arrive fully formed and that early incoherence is acceptable. Over time, people stop self-censoring embryonic ideas, and the organisation’s evolutionary capacity increases without any formal innovation programme.

4. Ask the recombination question

Evolutionary innovation rarely comes from creating something entirely new. It comes from recombining existing capabilities, assets, or insights in novel ways. Leaders can unlock this immediately by changing the dominant question. In any innovation or strategy discussion, ask once and only once: “What do we already have that could be used differently here?” Then stop talking.

This question redirects effort from invention to exaptation. It lowers cost, reduces risk, and accelerates movement into the adjacent possible. It also surfaces overlooked assets, underused data, and informal practices that rarely appear in formal plans. Practised consistently, this micro-practice shifts innovation from speculative brainstorming to practical evolution.

5. Name premature convergence when it happens

Groups often converge too quickly, especially in the presence of authority. Consensus can form not because an option is best, but because ambiguity feels uncomfortable and alignment feels safe. When agreement arrives unusually fast, intervene gently by naming the dynamic:

• “We’re converging quickly. What options might we be collapsing too early?”
• “What are we not exploring because this feels like the obvious answer?”

This is not about prolonging debate. It is about reopening variation long enough to ensure the system is not locking onto a suboptimal path. Naming convergence legitimises divergence without confrontation. It keeps decision quality high while preserving trust and momentum.

6. Harvest learning before killing an idea

In many organisations, evolutionary learning is lost because ideas are stopped without extraction. Failed experiments are discarded rather than mined, forcing the system to relearn the same lessons repeatedly.

Before ending an initiative, experiment, or pilot, pause and ask:

• “What did this teach us that we should reuse elsewhere?”
• “What capability, insight, or pattern emerged that still has value?”

Capture the learning explicitly, even if briefly, and move on. This practice ensures that variation produces retention, even when selection is negative. It reduces fear of experimentation, increases return on failed effort, and allows innovation to compound rather than reset. Over time, the organisation becomes more resilient because learning survives even when ideas do not.

This page is part of my broader work on complexity leadership, where I explore how leaders navigate uncertainty, sense patterns, and make decisions in complex systems.