by Kristian Hudson
I’ve long been drawn to systems thinking and complexity science—particularly how insights from complexity research could advance the field of implementation science. Complexity science invites us to embrace uncertainty, adaptiveness, and the interconnected nature of real-world systems—something I think is urgently needed in implementation.
Over the past few years, various research papers have caught my eye, especially those challenging linear “pipeline” models of implementation. For example, Greenhalgh and Papoutsi (2018) argue powerfully for a paradigm shift in health services research—one that embraces uncertainty, emergence, and dynamic interdependencies rather than trying to control or eliminate them. Similarly, Braithwaite et al. (2018) critique the traditional implementation model and call for approaches that reflect the non-linear, adaptive nature of health systems. Their work emphasizes the importance of feedback loops, local context, and the long timescales often required for meaningful change. Greenhalgh et al.’s NASSS framework (2017) also resonated deeply—it explores why certain innovations fail to scale or sustain, and how complexity at multiple levels (technological, organizational, systemic) shapes outcomes. (Check out a video I made on NASSS: YouTube link).
So, I was genuinely excited to record a podcast with Professor Harvey Maylor, Professor of Management Practice at Saïd Business School, University of Oxford. Harvey is one of the UK’s foremost experts in systems thinking, and I wanted to ask him: What does complexity really mean in practice? What gets in the way of implementation? And how can we shift systems toward better outcomes?”
What followed was one of the most insightful conversations I’ve had on the subject.
From Factory Floors to Government Policy
Harvey’s career began on the factory floor as a manufacturing engineer, where he quickly noticed that many operational problems weren’t local—they were systemic. Over time, he moved from examining the organisation of production lines to working with government departments and global firms to rethink how large-scale systems are structured and led.
As he put it:
“You can make big changes in how things are working, but there will always be a ceiling unless you can integrate with the next level up.”
It’s a sentiment many of us in implementation science know well. We may improve team communication, train new staff, or run an evidence-based intervention—only to see progress stall due to higher-level barriers: conflicting priorities, bureaucratic rules, and outdated reporting structures.
Harvey’s message was clear: implementation work doesn’t just need better interventions. It needs better systems of work.
Rules That Stop Progress
One of Harvey’s most powerful stories came from a road construction project that was racing to reopen before Christmas. Everything was ready to go—but for days, no work happened. Why? Because someone had decided that all new tasks must start on a Monday. That arbitrary rule led to a seven-day delay, derailed the project’s schedule, and cost a fortune.
This isn’t just about roads. Healthcare, education, and social care are full of similar constraints:
rules and routines that were never questioned but shape what’s possible. As Harvey noted:
“There are aspects of work design—rules, assumptions, ‘the way we do things’—that often turn out to be the biggest barriers to progress. And they’re usually within our power to change.”
Three Types of Complexity—and Why We Struggle with Two of Them
To make sense of what holds systems back, Harvey shared a model of three kinds of complexity:
- Structural Complexity – Related to the scale, pace, or regulation of work (e.g., how much needs to be done and how fast).
- Sociopolitical Complexity – Issues of power, relationships, communication, and trust.
- Emergent Complexity – Uncertainty and change; the stuff that can’t be planned, only responded to.
Most organizations focus heavily on the first. But it’s the second and third—how people relate to each other, and how they adapt to surprise—that tend to make or break implementation efforts. And yet, these areas are where we are least prepared.
Harvey argued that leadership needs to adapt accordingly:
- For structural issues, you need strong processes and systems.
- For sociopolitical issues, you need trust, listening, and co-production.
- For emergent complexity, you need entrepreneurial thinking and the courage to experiment.
The Power of Taking Ownership
One theme that kept coming up was agency. Systems can change—but only when someone takes ownership. Harvey told a story of a leadership team struggling with inefficiencies who kept blaming “the system.” Until, finally, someone said:
“Don’t you realise? We are the system.”
It’s easy to feel powerless inside large bureaucracies. But even in highly constrained environments, small experiments—changing a routine, redesigning a workflow, suspending a rule—can create openings for change. As Harvey put it:
“Often the system isn’t broken because of a lack of resources. It’s broken because of how the work is organised.”
The Case for Creating Slack
One of the most radical (and important) points Harvey made was this: the first step to improvement is often to unload the system.
Overloaded systems can’t adapt. When everything is maxed out—time, people, budgets—there’s no room for reflection, creativity, or experimentation. No slack means no space to grow.
“The paradox has to be understood: that simply putting more load on a system, doesn’t mean you get more out of it. And once we understand that, you can’t start to create a space that just might allow some of these changes to happen”.
This idea resonates deeply with what we’ve seen in implementation work, especially in the NHS. Systems can’t absorb change unless they have capacity to reflect and adapt. Without that, even the best-designed interventions fail.
Sustainably Self-Improving Systems
Harvey ended our conversation with a vision I’ll carry with me:
“A good goal is not short-term success. It’s a sustainably self-improving system.”
That means a system where improvement isn’t a one-off project or externally driven change—it’s just the way things are done. Where the people who do the work are empowered to change the work. Where feedback loops are built in. And where leadership fosters safety, trust, and experimentation.
As researchers and implementers, we can’t control all of this—but we can help create the conditions for it.
Final Thoughts
This conversation left me with a deeper appreciation for what complexity really means—and how we can work with it, not against it. Implementation science has made huge strides, but to fully embrace systems thinking, we need to go beyond logic models and fidelity checklists. We need to understand how systems behave under pressure, how people interact with rules, and how leadership can enable (or stifle) real change.
Harvey’s work—and his generous insights—are a reminder that meaningful improvement doesn’t come from doing more. It comes from doing things differently.
If you’re interested in listening to the full conversation, check out the full video here: https://www.youtube.com/watch?v=7esoqyAO6sI. In the meantime, I’d love to hear your thoughts: how are you seeing complexity in your own implementation work?
References
- Greenhalgh T, Papoutsi C. (2018). Studying complexity in health services research: desperately seeking an overdue paradigm shift. BMC Medicine, 16, 95.
- Braithwaite J, Churruca K, Long JC, Ellis LA, Herkes J. (2018). When complexity science meets implementation science: a theoretical and empirical analysis of systems change. BMC Medicine, 16, 63.
- Greenhalgh T, Wherton J, Papoutsi C, Lynch J, Hughes G, A’Court C, Hinder S, Fahy N, Procter R, Shaw S. (2017). Beyond adoption: a new framework for theorizing and evaluating nonadoption, abandonment, and challenges to the scale-up, spread, and sustainability of health and care technologies. J Med Internet Res, 19(11):e367.
