Systems Thinking for Product Managers: Systems Dynamics
How applying Systems Dynamics to Product Management can help you understand complex situations and causality.
In a previous article, I outlined some of the basic principles of Systems Thinking and how it can be used in Product Management.
Systems Dynamics
In this article, we will focus on Systems Dynamics, a Systems Thinking methodology. Touching briefly upon some key concepts and use cases to help introduce this to better to the Product community.
System Dynamics is a modeling approach used to understand and analyse complex systems over time and is part of Systems Theory. It involves the construction of feedback loops, stock-and-flow diagrams and simulations that represent the relationships and interactions within a system (a set of interconnecting things). It was initially developed in the 1950s to help managers improve their understanding of industrial processes. By applying System Dynamics, Product Managers can gain a deeper understanding of complex relationships that support product success, enabling Product Managers to create strategies that are robust and adaptable.
In the context of Product Management, Systems Dyanmics helps visualise and predict how changes in one part of the product ecosystem can ripple and affect the entire system (causation). It is particularly useful to interpret non-linear behaviour (nonlinearity) where small changes have disproportionally large effects. For example, a minor tweak in user interface design could lead to a significant increase in conversion rates such as when ASOS changed their checkout flow to allow users to checkout as guest, they were able to increase their conversion by 50%.
Process
To understand how to apply this methodology, you need to first know the process in which it can be applied. You do not need to complete all the stages in the process to gain insight, but they must be in sequence. Completing more stages will drive further insight, allowing for simulation and computational modelling.
Example casual loop diagram
You can read more the stages in the article ‘What is the system dynamics model’. Briefly, they are:
Define the problem: identify the problem or issue to be addressed, ensuring it is specific and actionable, and that you involve key stakeholders and understand the context.
Develop causal loop diagrams: identify key variables (nouns) that influence the problem, including both internal and external factors. Mapping these to create a diagram that shows the feedback and causal loops. Arrows to show direction of influence and indicate positive (+) or negative (-) relationships.
Develop ‘stock’ and ‘flow’ diagrams: staring from the causal loop diagram, include the stocks (accumulations of something over time) and flows (rates of change that affect the stocks). Rectangles to represent stocks and arrows to represent flows and include information on time delays and feedback loops.
Develop and test simulation models: using the causal loop and stock and flow diagrams, create computational models that simulate the behavior of the system over time. Test different scenarios and understand the impact of changes on the system.
Analyse results: analyse the outcome of the simulations, to identify patterns and insights. Considering the key drivers of behaviour.
Implement change and monitor: using the insights, develop action plans based on the findings, ensuring the plans are actionable. Once implemented, monitor and use feedback to assess its effectiveness so further adjustments can be made.
Bringing Casual Loops to Life
That’s a lot of theory! Let’s make this more practical/tangible by breaking this down using a very relatable scenario applying casual loop diagramming and including feedback loops and delays.
Imagine you are the Product Manager for a desk-booking application and it's doing reasonably well to compete with other products on the marketplace.
Recent research has indicated that users would like the ability to bulk-book a specific desk, i.e. the same desk every day for five days. You are unsure on whether to prioritise and pivot to deliver this feature over another (implementing Calendar view), which was planned into your roadmap and previous research also indicated that users would like.
Your product vision includes growing your market share as a major priority/focus.
You decide to apply Systems Dynamics to model the situation and your options, with the aim of determining the effect of changing priorities on your roadmap. You start with a causal loop diagram, identifying:
The key variables (as nouns), e.g. roadmap activities, prioritisation activities, etc.
The casual relationship between them.
Whether they are reinforcing or balancing
Whether there are any delays present
Causal loop diagram
Key components of a Casual Loop Diagram
Key concepts of a casual loop diagram:
Feedback Loops: are a core component of Systems Dynamics. They can be Reinforcing, (that is that have positive feedback and are represented by “R” or (+) in the diagram), or Balancing, (negative feedback represented by “B” or (-)). The relationship is positive if an increase in one variable also leads to an increase in the other, and if a decrease in one variable leads to a decrease in the other, i.e. both variables ‘go in the same direction’. The relationship is negative if an increase in one variable leads to a decrease in the other or vice versa, i.e. both variables ‘go in different directions’.
For example, a reinforcing loop might involve increasing customer satisfaction leading to more referrals, which in turn further increases customer satisfaction.
Delays: Delays refer to the time lag between an action and its visible impact.
For example, this could be the delay between investing in a new feature and seeing its effect on user engagement. Recognising delays helps to set realistic expectations and timelines and are denoted with two parallel lines (||).
Insights
Insights derived from the diagram and process:
Increased user research activities lead to more prioritisation and this in turn leads to further research activities, but there is a delay in user research having this effect. It is a reinforcing loop.
Increased roadmapping activities lead to further prioritisation and this in turn leads to further roadmapping. These activities create a reinforcing loop.
An increase in roadmapping activities can increase market share, and in turn, a change in market share increases roadmapping activities, a reinforcing loop.
There is a delay between user satisfaction and influence on market share as there is a delay before this has an effect. A decrease in user satisfaction will reduce market share, but user satisfaction is not influenced by market share, there is no reinforcing or balancing loop.
An increase in priorisation can decrease the need to feature pivot. However, feature pivots increase prioritisation. This creates a balancing loop
An increase in feature pivoting can decrease or increase user satisfaction, potentially causing further pivots, creating a reinforcing loop. Note there is a delay in influencing user satisfaction (implementation time) and a potential further pivot (time to analyse whether a further pivot is needed).
If we consider our scenario, this diagram could indicate that a feature pivot, when both features are considered of equal importance would be unwise, due to a reinforcing loop between pivoting and user satisfaction. This would also cause a ripple through into other activities. This diagram could be enhanced further, by adding causality between a feature pivot and impact on developers, technical debt and so on, leading to additional insight.
Example use cases
Further examples that could be applied to Product Management practice.
Feature prioritisation and roadmapping
Model to understand the broader implications of adding new features or modifying existing ones. By mapping the feedback loops between product features, user experience, and resource allocation, Product Managers can prioritise features that provide the greatest long-term value while avoiding those that could lead to negative outcomes like technical debt or user frustration.
Example: A team might model the impact of introducing a feature that requires substantial backend changes. This might show that while the feature could drive short-term growth, it also introduces complexity that slows down future development, ultimately harming the product in the long term.
Long-term strategic planning
System Dynamics helps to anticipate and plan for the cumulative effects of decisions. Instead of focusing only on immediate results, it helps to simulate how current actions will impact the product’s success over months or years, aligning short-term initiatives with long-term objectives.
Example: A team or service could explore the long-term impact of various growth strategies, e.g. expanding into new markets versus growth in existing markets. The model might show that rapid expansion leads to a temporary surge in user numbers but eventually strains customer support and infrastructure, causing a decline in service quality.
Considerations
As with many Systems Thinking approaches, this methodology can be extremely rewarding, but time consuming. I've found the biggest challenge to applying this approach is creating the time to see it through and gaining buy-in from stakeholders who may not initially appreciate its value. In addition, using the right tool can be pivotal, either complete the activity in person, or use a virtual whiteboarding product, e.g. Miro, Mural, etc.
Consider asking your team and stakeholders:
Do we understand the impact of changes we implement?
Do we understand or consider our interdependencies?
Are we confident in our strategic decisions?
If any of the answers to these questions is no, then Systems Dynamics can help to alleviate the negative impacts through exploration and exploring this space can be a compelling argument as to why the time is worth investing in this approach.
Conclusion
Product Management can involve managing dependencies, relationships and ensuring that change creates value. By taking the time to map out, explore and diagram through Systems Dynamics, teams can reduce the likelihood of negative consequences and in turn, enhance positive feedback loops.
This methodology enables a holistic approach to decision making, avoiding the pitfalls of traditional linear approaches. It helps Product Managers to view what goal or objective will be a reinforcing factor within the product strategy whilst avoiding balancing (negative) feedback loops.
Further reading:
Systems Thinking for Product Managers (to be continued..)
This is the first article in a series I'll release within Product Breaks, focusing on Systems Thinking methodologies and Product Management. Further articles will aim to cover the topics:
Critical Systems Heuristics (CSH)
Viable System Model (VSM)
Soft Systems Methodology (SSM)
Strategic Options Development and Analysis (SODA)
 | A guest post by
|