A step-by-

step guide to
Systems
Mapping

Retrieved from: https://xynteo.com/tools/wicked-problems-toolkit

01 | How system mapping can tackle
wicked problems?

By using system mapping to understand wicked problem systems in their

entirety, it’s possible to find new ways of tackling these problems e.g. by
understanding why the system is behaving the way it is, and seeing
connections that are not obvious.

From there, you can develop smart interventions that address the
multiple root causes of wicked problems.

Why are we talking about ‘interventions’?

To start practicing systems thinking, we must shift from thinking about

‘solutions’ to thinking about ‘interventions’.

“The first step is to shift the goal of action on significant problems from
‘solution’ to ‘intervention’. Instead of seeking the answer that totally
eliminates a problem, one should recognise that actions occur in an
ongoing process, and further actions will always be needed.”

‘Wholesome Design for Wicked Problems’, Robert Knapp.

1
02 | When to use system mapping

• When you want to build a shared, holistic understanding of a multi-

dimensional problem (especially to bring together perspectives from
different stakeholders).

• When you want to identify unexpected connections and gaps in

your understanding of a problem system.

• When you want to make smart choices about how to intervene

effectively in a system for maximum impact.

• When you want to mitigate the risks of unintended consequences of

an action.

• When you want to reduce the risk of developing superficial

solutions – ones that may act as a temporary, short-term fix but fail
to solve problems.

• When you have designed a solution, but it is working in unexpected

or counterintuitive ways, and you want to understand it better.

2
03 | Stage 1: Prepare to map your System

Step 1
Writing your Problem statement

Step 2
Setting your System boundary

Step 3
Researching System dynamics

Step 4
Conducting Stakeholder interviews

3
04 | Writing your problem statement

To start mapping a problem system, you should first be clear about what
the problem is. Writing a short ‘problem statement’ can provide this
clarity.

A good problem statement will explain what the problem is, how it is
manifesting today, who is affected by it, and which other problems it
perpetuates. (NOTE: you don’t have to describe why it persists – that’s for
your system map to answer!)

You can choose to end your problem statement with a ‘how might we…’
question that helps frame the problem in terms of an aspirational and
actionable goal. ( NOTE: be careful not to bake a solution into your ‘how
might we...’ statement – it’s important to remain open-minded
about possible interventions!)

Example of a problem Statement

One in six people lack access to electricity. Most are poor and live in rural
areas. One in two also lack access to clean cooking facilities. These people
either use dirty or polluting fuels, or spend a lot of time collecting fuel to
meet basic needs. This has a negative impact on their health and their
ability to progress on other fronts, such as education and employment.
How might we provide clean, reliable and affordable energy to the rural
poor?
4
05 | Setting your system boundary

Once you’ve written your problem statement, it’s time to start scoping
your system map.

The system boundary may be obvious within your problem statement,

or you may have to make a decision on a starting point. This might be a
geographical focus (e.g. East Africa), or a specific part of a system e.g.
upstream food waste at the farm, vs downstream food waste by
consumers.

It’s always possible to adjust your boundary later, but it is useful to draw
some limits now to help you narrow in on the most fundamental
components in your system.
5
06 | Researching system dynamics
Once you’ve scoped your system boundary, it’s time to get googling. Pull out
articles, academic papers, reports, interviews – anything that contains interesting
insights on your problem system.
The aim of this step is to gather as many ‘system dynamics’ as possible within your
chosen system.
A system dynamic is any form of causal relationship that exists within a system. If X
causes Y, it’s a system dynamic.
If A results from B, it’s a system dynamic. As you read through your sources, highlight
causality wherever you see it – this is a quick and easy way to start drawing out system
dynamics.
Let’s say your chosen problem system is ‘rising obesity levels’ and you’ve scoped your
system boundary further to focus on rising obesity in India.
Here’s an excerpt that contains some interesting system dynamics: can you identify
them?

India's obesity crisis

Prakash Shetty, of the Food and Agriculture Organisation (FAO), says lifestyle and food
habits change as an economy develops. There is a significant increase in the consumption
of fats, sugars and energy-dense foods. The main drivers behind changing lifestyle and
food habits are rising incomes, urbanisation and globalisation.
Rising incomes and urbanisation lead to the substitution of servants or appliances for
physical work around the house, while family breadwinners take to desk jobs instead of
ploughing the fields. These factors also encourage more sedentary pursuits such as
television viewing and computer use, and well-off city-dwellers travel by car instead of
walking or cycling.
At the same time, globalisation puts junk food and fast food within easy reach of a
population often hard-pressed to find time to cook healthy meals, but with more
than enough money to buy a greasy lunch at a nearby restaurant.
In India, these factors have contributed to the rise of bad eating habits and lack of
exercise amongst a growing urban middle class, and their effects are startlingly visible.
A University of North Carolina study conducted in Andhra Pradesh showed that 37
percent of women living in cities are clinically overweight or obese, and a study by the All
India Institute of Medical Sciences (AIIMS) found that 76 percent of women in Delhi
suffer from abdominal obesity.
But India’s obesity crisis has been further exacerbated by some peculiarly Indian factors.
Despite India being a vegetarianism stronghold, the average Indian’s diet severely
lacks fruit and vegetables – an abysmal low of 150 gm a day against the WHO
recommendation of 400 gm a day.
Instead, Indian consumers’ calories come mostly from refined carbohydrates and fats.
Indian food also tends to be amongst the most oil-rich in the world, and while Indians
have taken Western junk food to heart (quite literally!), India also has its own array of
fried snacks (think pakoras and samosas) which make the population susceptible to
weight gain without even factoring in globalisation.
To add to this dismal picture are schools which promote a culture of fierce academic
competitiveness and leave children with no time to play or exercise. That Indians are as
an ethnic group more genetically prone to obesity and its health consequences is the
cherry on top. Indians have up to five percent more body fat than Caucasians at a given
weight and height, suggesting that Indians may face higher health risks than expected at
a given weight. Some researchers have even gone so far as to suggest that Indians are
genetically more likely to store fat due to a ‘thrifty gene’ that evolved out of
undernourishment in the past.
6
07 | Conducting stakeholder interviews

Researching online is a great place to start, but to really flesh out the
nuances within your system map you have to speak to some people who
have been deeply immersed in your problem system: either because they are
professionally or academically focused on it, or they are actually living it.

The aim of this step is to gather as much insight as possible on system dynamics
within your chosen system from a diverse set of stakeholders.

Who should I speak to?

Try to get a diverse set of perspectives on your problem system. Choose people
who are operating in different parts of the system e.g. upstream and
downstream, directly involved and peripherally involved.

How many conversations should I aim for?

Consider the ‘rule of six’: you can get over 90% of the insight you need to make an
effective system map by having six good conversations. After that, the incremental
insight per conversation reduces significantly.
You can get away with fewer than six stakeholder interviews if you are time-
constrained, but you also don’t need to go well above six unless you have good
reason.

What should I ask?

Come prepared with a set of questions, but be prepared to let your interviewee
steer the conversation. As long as it’s related to your system and the dynamics
within it, it should be interesting to you – by keeping an open mind, you
may chance upon some hidden insight.

7
08 | Stage 2: Build your System Map

Step 1
Mapping your core nodes

Step 2
Drawing dynamics and adding nodes

Step 3
Identifying feedback loops

Step 4
Testing and refining your system map

8
09 | About nodes

Before mapping your core nodes, it’s worth learning what a node is.

In simple terms, a node is something that can influence/be influenced by

something else. Or in ‘systems speak’ – a node is the outcome of an influencing
dynamic. Confused? Think back to the system dynamics you identified on card 27
and card 28 – where X leads to Y, or A causes B. X, Y, A and B are all nodes.

Here are some examples of nodes in the obesity problem system. Pay special
attention to the phrasing:

Common confusions:

• It’s worth noting that nodes are not just nouns – ideally they are things that
can have a measurable quality/quantity. You might want to augment
or diminish the quality/quantity.
• For instance, ‘refugees’ is not a node – that’s just a group of stakeholders.
What about them? ‘Number of refugees employed’ or ‘morale of recent
refugees’ are both node-worthy: they can influence and be influenced.

9
10 | Mapping your core nodes

When you start a system map, your first step is to lay out the core nodes in your
system.

Start with things that show up in your problem statement – a core node will likely
be pivotal to the rest of the dynamics in your system.

Starting from this basic framework, you’ll be able to add additional nodes and
draw out the dynamics between them.

10
11 | About dynamics

Dynamics describe how one node influences another: in other words, any
causal relationship within your system is a dynamic. Note that one node
can influence another in two ways – positive and negative – and the effect
can be immediate, or delayed:

Positive influence

More of A leads to more of B

Examples:
Prevalence of mobile phones leads to more access to pricing information for
smallholder farmers.
Spending on advertising leads to more sales.

Negative influence

More of A leads to less of B

Examples:
Rainy days in summer lead to lower sales of ice cream.
Spending on software leads to less time spent on billing customers.

Delays

A changes B, but after some time

Examples:
Malnutrition during childhood leads to mental stunting in adults.
A decline in sales this month will be reflected in inventories next month.
Common confusions:
• All dynamics are either positive or negative – and the delay just adds a
timescale modifier.
• Positive doesn’t mean ‘good’ and negative doesn’t mean ‘bad’ – it just
indicates the direction of the influence. 11
12 | Drawing dynamics and adding nodes

It’s time now to begin adding dynamics between the core nodes on your
system map.

• Remember to think of a dynamic as ‘more of X leads to more of Y’.

• When connecting nodes, make sure you have covered the logical steps
between them.
For instance: lack of education doesn’t directly lead to poor health outcomes.
There are steps in between (e.g. lower-paid jobs, lower income, etc.).
• Add additional nodes to represent other elements in the system and increase
resolution.
• To add nodes, start by looking at an existing node and asking the simple
question – what causes this? That will be your next node. Then ask – what
results from this? That’s another node.
12
13 | Identifying feedback loops

When positive influences, negative influences, and delays combine,

systems start to exhibit behaviours that are familiar to us in the real world.
Two simple examples are reinforcing and stabilising loops. Look for some of
these patterns on your system map; you may choose to circle or label
them to come back to them later.

Reinforcing feedback loop

+ More sales of a good product leads to more satisfied customers.

+ More satisfied customers lead to more positive word-of-mouth to other
potential customers.
+ With some delay, this additional positive word-of-mouth leads to increased
sales.

Stabilising feedback loop

+ More immigrants entering a country leads to greater fears around

immigrants taking jobs.
+ After a delay, this leads to pressure on government to pass stricter laws aimed
at slowing down immigration.
- This reduces overall numbers of immigrants entering the country. 13
14 | Testing and refining your map

Having made a first pass at a system map, now is the time to test and
refine it.
Review it:
Does it help to answer the kinds of questions that you are trying to answer?
Test it:
Test it offline with stakeholders and ensure that the dynamics are captured
correctly, so that their narratives and your map reinforce each other.
Refine it:
You may need to adjust the boundaries of the system, so that some of what was
in the ‘outside world’ is included. And you may need to add or rewrite nodes, or
tweak dynamics.
Important:
The aim is not to get to a system map that is perfect – just to create something
that is good enough to start working with.
14
15 | Stage 3: Identify points of intervention

Step 1
Determining points of intervention

Step 2
Prioritising points of intervention

15
16 | Determining points of intervention

Now that you have built out your system map, it’s time to identify points of
intervention: areas where you think taking action would have a positive
effect through the system.

Where on your map would a change in the value of a node produce a

beneficial ripple effect across the system? Experiment with different nodes
and follow the effects along the dynamics on the map.

You may choose to mark potential points of intervention on your map by

circling nodes, feedback loops or larger sections of your map.

16
17 | Prioritizing points of intervention

Now you need to decide which of these points of intervention to pursue.

Where do you have the ability to make the needed impact? Consider
filtering your options using the following criteria:

1. Is the impact likely to have a positive domino effect through the

wider system?

2. Do you, your team or your wider network have the ability to trigger
the proposed change?

3. Have others intervened in a similar way? What can you learn from
the results of their actions?

4. Is it worth taking action at multiple points of intervention?

Identifying points of intervention is only the first step – following this, you
will need to design and deploy these interventions effectively to have the
system effect you
are seeking.

17
18 | System map examples

Waste problem in Mumbai

Deforestation in Brazil

18