Systems thinking helps organisations understand the wider structures shaping behaviour, decision-making and outcomes. Rather than focusing only on individual actions, systems thinking looks at the processes, incentives, relationships and environmental factors influencing behaviour across the wider system.

In our first article, we looked at where to start with systems mapping. We shared three different systems-mapping tools, including process mapping, power mapping and problem mapping, and explored how to zoom out, map the system, dig beneath the surface, and identifywhere effort is most likely to make a difference.

That first step of mapping matters because more activity does not always mean more impact. More spend, better targeting and sharper messaging can still fail to move the needle if the real barrier sits elsewhere in the system.

Systems thinking asks a different question:  
Not just “How do we influence behaviour?”  
But “What is shaping the behaviour in the first place?”

But once you’ve mapped the system, the next question is obvious:

What do we do now?

This is where systems thinking becomes practical.

Remove friction from the process

A good place to start is removing any major pain points and barriers across the process. Process maps extend beyond traditional customer journey maps by showing the organisational back-end processes sitting behind the experience. You can think of them as showing both the front and back stages of a system.

A customer may understand the message and want the outcome, but the process makes it harder than it needs to be. There may be too many steps, unclear handoffs, poorly timed communications, confusing rules, compliance requirements, default settings, or moments where uncertainty builds.

Systems thinking helps identify leverage points: small shifts that can create disproportionate change across the wider system.

The solution does not always need to be a complete process overhaul. It might be changing a default, removing one unnecessary step, shifting the timing of communication, improving a handoff, or providing support when people get stuck.

Small changes can create disproportionate impact because friction compounds across the wider system.

If the map shows friction built into the process, the solution needs to work on the process, not just the message.

Using power mapping to identify influence

If process mapping tells us what to focus on, power mapping reveals who actually shapes the outcome.

The person you are targeting is not always the person with the most influence over the behaviour. People may be more influenced by a peer, family member, an industry body or cultural signal than by the organisation running the campaign.

At The Research Agency (TRA), we call this ‘supporting the supporters’ or ‘influencing the influencers’. We have a guide on tapping into this messenger effect, because we know it’s not just about the message – it’s also about who it’s from.  

Instead of asking “What should we say?”, ask:

Who does our audience connect with?
Who does our audience trust?
Who has the biggest influence over the system?
Who can make the behaviour easier, safer or more normal?
Who is benefiting from the current system?
Who could help shift it?

Power mapping can point toward new messengers, new partnerships, new incentives or new permissions.

Sometimes the organisation trying to create change is not the right messenger at all. Systems thinking helps uncover where trust already exists, so organisations can work with it rather than against it.

What is reinforcing the current behaviour?

Systems keep producing the same outcomes for a reason.

Current behaviours are often reinforced by the rules, incentives, measures, norms or rewards sitting around them. In systems thinking, these are often described as reinforcing loops or causal loops: patterns where one part of the system keeps feeding another.

If those reinforcers stay the same, the system often pulls people back to the same behaviour, even after a well-designed intervention.

Ask:

What gets rewarded?
What gets measured?
What gets normalised?
What makes the current behaviour easier than the desired one?
What keeps pulling the system back to the current state?

If teams are told to collaborate, but scorecards reward department-level performance, then the system is reinforcing silos. If customers are encouraged to make a change, but the status quo option requires less effort and feels more rewarding, then the system is reinforcing inertia.

Identifying the reinforcer helps us understand where to focus our energy and effort.

What problem are you actually solving?

A systems map should help you define and understand the type of problem you are dealing with.

Firstly, the problem mapping framework or iceberg model helps us understand if we’re dealing with a deep-seated belief or mindset problem, or if it’s a structural and process-based problem.  

The visible behaviour is often just the tip of the iceberg. Beneath it sit patterns, structures, incentives and mental models shaping what people do.

In addition to problem mapping or the iceberg model, the Cynefin model is also useful. It helps distinguish between clear, complicated, complex and chaotic problems, and each requires a different response.  

A clear problem has an established answer. The solution is usually to apply best practice, simplify the process or remove obvious friction.

A complicated problem has an answer, but it needs expertise. It will likely require technical analysis, specialist input or careful design.

A complex problem has many interacting factors, and the outcome can’t be fully predicted upfront. The solution is likely to involve testing, learning and adapting over time.

A chaotic problem is unstable and unclear. Leaders usually need to act first, sense what is happening, then respond, often with a novel approach to create enough stability to move forward.

For practitioners, this is a useful discipline. Not every challenge needs a full system transformation. But not every challenge can be solved with another message either.

Reframe the problem

How a problem is framed shapes the solutions that become visible.

If low uptake is framed as an awareness problem, the likely answer is more communication. If it is framed as a trust problem, the answer may be a different messenger. If it is framed as a process problem, the answer may be a process shift. If it is framed as a market problem, the answer may sit with suppliers, retailers, incentives or regulation.

This is why problem mapping matters.

It helps separate what is visible from what is driving the issue underneath. Sometimes the solution is not to respond to the visible behaviour, but to reframe the challenge itself.

Waste minimisation is one problem. Making new products from waste is another. Both relate to the same issue, but they open up very different solutions, partnerships and measures of success.

The question is: what changes if we describe the problem differently?

Sometimes reframing the problem changes the types of solutions that become possible.

Test, measure, learn and adapt

In complex systems, solutions are rarely certain upfront.

When many actors, incentives and behaviours are interacting, the goal is not to design the perfect intervention on paper. The goal is to test, measure, learn and adapt.

If process mapping shows pressure points in the customer journey, those moments can become test sites, especially if your organisation is mapping customer data across the key journey stages. For example, if VOC data or call centre measures show confusion at a particular stage, you can pilot a process change, measure whether the issue reduces, then adapt.

Likewise, if power mapping points to a new partnership, you can measure whether that partnership increases the desired outcome, whether that is uptake, confidence, completion, conversion or compliance.

The point is to watch how the system responds.

What changed?
What stayed the same?
What adapted around the intervention?
What did we learn about where the leverage really sits?

Turning system maps into action

Your map should change what you do.

Problem mapping helps you understand what type of issue you are dealing with, whether the problem needs reframing, and how deep the solution needs to go.

Process mapping helps you remove friction, adjust moments and challenge the process boundary.

Power mapping helps you find trusted messengers, recruit allies, shift incentives and challenge the influence boundary.

And if the goal is systemic change, there will rarely be one single solution. More often, change comes from multiple levers working together: a better process, a different messenger, a reframed problem, a changed incentive, and a portfolio of tests.

The point is not to stop doing good marketing, CX or behaviour change. It is to make sure those efforts are aimed at the right part of the system.

Zoom out.
Dig deep.
Focus effort.
Test, measure and adapt.

Because nudges can treat symptoms. Systems thinking helps find the structure shaping behaviour, and the levers that can shift it.

Watch our systems thinking webinar on demand, or explore more behavioural science insights, frameworks, and research in our Insights Hub.

‍