Navigating Complexity: Unraveling the causal structure of real-world systems
Part Two: Bioform models can help uncover the nuanced causal structure of organizations— and adapt as they evolve.
- Part One: Understand Uncertainty: From simple models to complex systems
- Part Three: Bioform models: So, what?
Introduction
Imagine you’re the captain of a ship navigating through a vast and unpredictable sea. Each member of your crew has a unique role and contributes to the overall journey in their own way. Your task is to understand these roles and interactions to ensure the smooth sailing of your ship. This is not unlike managing an organization, where numerous factors interact in complex ways to create outcomes and accomplish goals. Welcome to the world of Bioform models.
In the first part of our series, we introduced bioform models and methodology, a novel approach that infers — rather than posits — the causal structure of a system, going beyond mere pattern recognition to capture the underlying dynamics that generate these patterns.
Now, we delve deeper into the realm of real-world complexity: companies*. We’ll show you how Bioform models can identify the drivers of a company’s revenue, paving the way for our third piece in the series where we’ll explore what this means for leaders of organizations of different types.
*It’s important to note: We at Bioform Labs are working on a broad tapestry of applications of these models: ecosystem stewardship, nature-backed currencies, economic development, public health, organizational resilience, product strategy and growth, and others. We use the example of a company to illustrate the potential and near-term applications of our toolkit in a familiar context. If you are interested in how these can apply to your organization or field of study — reach out!
Bioform Models in Action: Decoding Revenue Drivers
Companies are complex systems, with numerous interconnected parts influencing each other in non-linear and often unpredictable ways. To truly understand these systems, we need a methodology capable of capturing these dynamics. This is where Bioform models come in.
In a recent project, we used Bioform models to decode the drivers of a company’s revenue. The results were illuminating. We found that financial efficiency and investment strategy emerged as significant predictors of revenue. However, other factors such as operational efficiency and market dynamics also played crucial roles. This nuanced causal structure, revealed by the Bioform models, provides a more accurate and comprehensive understanding of the company’s revenue dynamics, as observed through financial and economic data.
Indeed, while some of these findings may seem like common sense — the importance of financial efficiency or investment strategy, for instance — it’s crucial to remember the way we arrived at these insights.
Our Bioform model doesn’t come pre-loaded with any understanding of business, markets, customers, or what might drive revenue. It isn’t designed to automatically favor certain variables over others or biased by preconceived notions of what ‘should’ matter. Instead, it functions as a kind of ‘causal compass,’ directing us toward potential cause-and-effect relationships within a vast sea of data.
We started with an incredibly limited dataset and a model that, at the outset, knew nothing about the company or its business environment. From there, we successfully derived causal models that indicate what’s driving this specific company’s revenue. This didn’t require supercomputers or vast computational resources — just a methodical, data-driven exploration of potential causal structures.
Furthermore, the resulting models aren’t just academic exercises. They have practical implications, offering insights that can guide the decision-making of executives, managers, and analysts.
Understanding Organizations with Bioform Models
Now, let’s take a step back and look at how we arrived at these insights. We started by defining the system of interest — the company’s revenue generation process — and identifying the main factor we wanted to predict — the company’s revenue. We used publicly available data about the company and the economy, spanning roughly three decades.
Our approach was to create a functional representation of the company’s revenue generation process. We chose variables that we believed could have a significant impact on the company’s revenue, based on our understanding of the company and its industry. These included financial efficiency, investment strategy, and market dynamics. We then used Bioform models to explore how these variables interacted with each other and contributed to revenue generation. This process allowed us to iteratively refine our understanding of the system, leading to a more accurate and nuanced model of the company’s revenue dynamics.
This iterative process of refinement is a key aspect of Bioform modeling. It’s not a one-and-done approach, but a continuous exploration of potential causal structures, each time refining our hypothesis based on the results and attempting to increase our model’s ability to accurately predict revenue in different states and conditions.
We grouped the variables into different categories we could explore as potential drivers of revenue. These categories formed the basis for the types of models we would build. They included operational efficiency (e.g., revenue collection), financial efficiency (e.g., asset turnover), investment strategy (e.g., capital expenditure), and market dynamics (e.g., investor sentiment, consumer optimism, fed funds rate).
But how do we measure the success of these models? We use a metric called Variational Free Energy (VFE) level. VFE is a measure of how well the model’s predictions match the observed data. A low VFE indicates low “surprise”, meaning the model’s predictions closely align with reality, thus indicating high predictability and accuracy.
In total, we built and tested 23 models. The variability in VFE levels and volatility across models points to the complexity of the causal structure and indicates potential areas for further investigation.
Model Performance and Variability
In our analysis, we found that the relevance and predictive power of the models varied over time and across different scenarios, reflecting the dynamic nature of the company and its environment. Models focusing on financial efficiency and investment strategy, including variables like Total Debt Equity and Capital Expenditure, showed high accuracy but their performance varied across different time periods and event spaces. This suggests that while these factors are significant drivers of a company’s revenue, their influence may fluctuate depending on the company’s specific circumstances at any given time.
On the other hand, models capturing comprehensive market dynamics, such as the Fed Funds Rate, Consumer Confidence, and Institutional Investor Confidence, were more stable but less accurate. This indicates that while these factors do impact a company’s revenue, they may not be the primary drivers. Moreover, their influence appears to be more consistent over time, suggesting that they represent broader economic trends that affect all companies, not just the one we were studying.
This variability in model performance underscores the complexity of the causal structure and the dynamic nature of real-world systems. It highlights the fact that there is no one-size-fits-all model that can accurately predict a company’s revenue across all time periods and event spaces. Instead, the most effective approach is to use a range of models, each tailored to different scenarios and time periods, and to continually update and refine these models based on new data and insights. This iterative process of model building, testing, and refinement is a key aspect of Bioform modeling and is essential for navigating the complex and ever-changing landscape of business.
Unraveling Complexity: Like Navigating a Stormy Sea
To make this process more tangible, let’s expand on the metaphor from above. Imagine you’re the captain of a ship trying to navigate through a stormy sea. You have a general idea of the direction you need to go, and you can observe the current conditions to understand your immediate situation.
Applying Bioform models in this scenario is like starting with a basic course and then refining it based on the changing conditions. You set a course, observe the effects of the wind and waves, and then adjust your course based on your observations. Maybe you steer more to the east if the wind is pushing you west, or perhaps you slow down if the waves are too high. You continue this process of setting a course, observing, and adjusting until you reach your destination.
In the context of Bioform models, the ‘destination’ is the revenue we want to predict, the ‘course’ is the model we’re using, and the ‘wind and waves’ are the variables we believe affect this revenue. The ‘setting a course, observing, and adjusting’ is the process of testing and refining our model. If the model’s predictions aren’t accurate enough (we’re off course), we revisit our ‘navigation plan’ (the model), adjust our ‘course’ (variables), and try again. This iterative process of building, testing, and refining a Bioform model mirrors the navigation process in our metaphor, underscoring the dynamic and adaptive nature of our approach.
The Dynamic Nature of Real World Systems
This metaphor underscores the iterative and adaptive nature of our work at Bioform Labs. We recognize that complex systems like how a company generates revenue aren’t static; they’re dynamic and constantly evolving, much like a ship navigating through a stormy sea. Our models reflect this dynamism, continually adapting and refining their predictions based on new data and insights.
Moreover, we understand that the relevance and predictive power of these models can vary over time and across different scenarios. Just as a ship’s course may need to be adjusted based on changing weather conditions or unexpected obstacles, our models may need to be adjusted based on changes in the company’s environment or business strategy.
This recognition of the dynamic nature of real-world systems and the need for models that can adapt to these changes is a key aspect of Bioform models. It sets them apart from traditional analytical techniques, which often treat systems as static and unchanging. Instead of trying to find a one-size-fits-all model, we embrace the complexity and variability of the systems we study, using a range of models tailored to different scenarios and continually refining these models based on new data and insights. This approach allows us to navigate the complex and ever-changing landscape of business with greater accuracy and insight.
So, while these findings might align with what some might call ‘common sense,’ the way we got here is anything but. It’s a testament to the power of empirical, model-based exploration and the valuable insights that this approach can generate.
In this exploration, we’ve seen how Bioform models offer a unique approach to understanding complex systems. They allow us to navigate the stormy seas of complexity and uncertainty, guiding us towards a deeper understanding of the systems we’re studying. Whether we’re predicting a company’s revenue or modeling the dynamics of a market, product or ecosystem, Bioform models provide a powerful tool for unraveling complexity and making sense of the world around us.
We’ve demonstrated how these models can decode the drivers of a company’s revenue, revealing a nuanced causal structure that traditional analysis methods might miss. This ability to capture the dynamic and interconnected nature of real-world systems sets Bioform models apart, offering valuable insights that can guide strategic decision-making and resource allocation.
In the next and final post in this series, we’ll explore how Bioform models can comprise a 21st-century toolkit that is critical for organizations of all kinds — whether commercial, social, or environmentally driven — to improve their ability to adapt…and thus enhance resilience, productivity, and impact.
Continue to Part Three: Bioform models: So, what?
Sign-up for Early Access
We’re excited to have you along on the journey, and we invite you to be part of it. By signing up for early access to our platform, you’ll be among the first to experience the power of Bioform models. You’ll gain a new perspective on the complex systems that drive your business or field of study, and you’ll have the opportunity to explore these systems in a way that’s both intuitive and grounded in data.
If you’re interested in navigating the complexities of our uncertain world with us, we invite you to sign up for early access to our platform at: https://www.bioformlabs.org
And if you’re curious about how Bioform models can help address a specific challenge or need, like uncovering unexpected causal relationships to drive growth, we’d love to have a discussion. Reach out to joshua@bioformlabs.org or cory@bioformlabs.org. Let’s explore together how Bioform models can provide the insights you need to navigate your unique challenges and opportunities.
