Systems Thinking in Project Management: Lessons from Donella Meadows

Donella Meadows, the systems thinker behind Thinking in Systems, died in 2001 before her manuscript was finished. Her colleagues and students edited, structured and published it in 2008.

That's right, the woman who wrote about systems never saw the final book! Yet the system around her – relationships, routines, shared language and purpose – was strong enough to carry the work.

For leaders, that raises a confronting question: if you stepped away tomorrow, would your current project management system (think purpose, people, process) finish the work well, or would it stall without you?

Why systems thinking matters in project management

Systems are more than the sum of their parts, and it is often difficult to see and design for positive outcomes for the whole.

In projects we tend to see the world as a string of events: the crisis steering meeting, the messy cut-over weekend, the unexpected server fail, the “heroic” recovery that leaves everyone exhausted. Meadows pushes us to lift our gaze from discrete events to system behaviour and structure.

The challenge is the same across retail operations and during any ERP system implementation: designing systems that continue to work when conditions inevitably change.

Three system characteristics matter especially for projects:

1. Resilience: the ability of the project system to absorb shocks and shifts and still achieve its purpose.

2. Self-organisation: the capacity of people and processes to re-configure themselves intelligently as reality changes.

3. Hierarchy: the structure of decision-making and accountability between parts and whole.

When we treat a project as a series of tasks rather than a living system with these characteristics, we overestimate our control and under-estimate the second- and third-order consequences of our decisions.

We are more ignorant than we like to admit

One of Meadows' most useful provocations is her line that "we are terrifyingly ignorant." She reminds us that:

• Everything we think we know about the world is a model.

• Our models often have strong congruence with the world.

• Our models still fall short of the world's full complexity.

In project work, that ignorance shows up every time we are surprised by the impact of a change request, a project governance tweak or a new incentive.

We add an approval step to "reduce risk" and discover we have created delay, frustration and shadow workarounds. We tie bonuses to go-live dates and discover people will hit the date at the cost of quality, rework and trust.

PMI's research on systems thinking and project outcomes highlights that projects in complex environments perform better when feedback loops and interdependencies are explicitly considered.

Harvard Business Review has repeatedly shown how misaligned economic incentives backfire, undermining the behaviours leaders hope to encourage. (HBR, 2009, 2024.)

When people lose a sense of responsibility and intrinsic motivation, performance becomes brittle and dependent on external controls.

The tension for retail leaders is that you need enough structure to steer investment and risk, without suffocating the self-organising ability that makes effective change management possible. That balance sits at the heart of organisational change management.

What we're learning from this

At 6R, the more we sit alongside teams through implementations, the more we see projects as systems within systems.

The project exists inside the business system, and each influences the other. Designing or correcting those systems so they produce outcomes that are genuinely desirable is hard work during any system implementation.

Here's what we're putting into practice as we apply Meadows' lens of resilience, self-organisation and hierarchy.

Organisational resilience: beyond contingency plans

Organisational resilience is not just risk logs and contingency plans. It's the capacity of the project management system to absorb shocks – staff turnover, supplier delays, bad data, shifting scope – and still orient towards its purpose.

We see more resilience when:

• Time horizons extend beyond go-live to three- to five-year impact, so decisions are made with long-term consequences in mind.

• People are trusted with intrinsic responsibility, not overridden with layers of inspections and controls.

• The customer feedback loop is designed deliberately: issues at the customer layer surface quickly and change (hopefully) how decisions are made, rather than being flattened into summaries and averages.

When resilience is low, big important work gets consistently postponed in favour of short-term band-aids and hodge-podge fixes. The project becomes fragile: every surprise demands a crisis meeting instead of being absorbed by a flexible, self-correcting system.

“A resilient project system doesn't eliminate shocks; it makes sure shocks don't decide the outcome.”

Self-organisation: trusting the people in the system

Self-organisation is the ability of a system to re-configure its own structure as conditions change.

Meadows treats this as a high-leverage characteristic, sitting close to her upper leverage points: the power to add, change or evolve system structure, the goals of the system and the paradigm it arises from.

Applied to projects, we are learning to ask:

• Where can the teams closest to the work change process, instead of waiting for head office sign-off?

• Where are rules so rigid that intelligent local adaptation is punished rather than encouraged?

• Where do our own behaviours as leaders signal that "thinking for yourself" is risky?

When self-organisation is supported, the project gains thousands of small experiments and corrections. When it is suppressed, the only "adaptation" happens in the shadows – unofficial spreadsheets, side chats and unlogged decisions.

Hierarchy: balancing part and whole

Somebody has to make calls, set direction, and hold accountability.

Meadows' insight is that hierarchy must stay in balance between optimising subsystems and optimising the whole. Too much control from the top pushes too far towards 'average' rules at the expense of local intelligence.

Across retail project management, we are paying more attention to:

• Whether decision rights match responsibility – are people accountable for outcomes they don't control?

• Whether project governance is there to learn and adjust or approve and police.

• Whether subsystems (stores, DCs, e-commerce, finance) are rewarded for behaviour that erodes whole-of-business performance – a project-scale tragedy of the commons.

Balancing hierarchy means giving the parts enough freedom to self-organise within a clear purpose and shared rules that protect the whole.

A practical lens: Meadows' leverage points for retail project management

Meadows lists twelve places to intervene in a system, from the least impactful (parameters and numbers) to the most impactful (paradigms and the power to transcend them). When we apply this in projects, we notice how much energy goes into the lower leverage points:

• More reports and dashboards (information flows)

• More approvals and RACI matrices (rules)

• More KPIs and bonus schemes (numbers)

Far less energy goes into the higher leverage points:

• Redefining the goals of the project system in long-term, whole-of-business terms.

• Challenging the mindset that "speed and savings now" is always the right answer.

• Building cultures where people can question the paradigm, not just work within it.

The greatest leverage comes from changing the system rather than simply adding more controls.

The hidden mechanism many leaders miss is that changing the rules, goals and paradigms of a project management system has far more impact than any single technical decision.

Retail improvement, made practical.
Leadership thinking that drives change.

Sign up to receive new articles and strategic guidance.