What is it about?

This research looks at how businesses form connections and collaborate with each other, focusing on how decision-making happens in real life. Instead of assuming that businesses always act rationally, we model them using rules similar to human behavior—what we call heuristics. For instance, when companies decide whether to collaborate or compete, they don't always base their choices on perfect logic. Instead, their decisions may come from past experiences, personal biases, or limited information, much like how people make choices in day-to-day life. Similarly, when they choose to connect, they do not necessarily conduct a thorough search but choose the first good-enough option. This adds to modeling complexity but also allows us to see what happens when we include human characteristics in the form of propensities to connect and propensities to cooperate, along with imperfect updates. So, in this research, we draw from network science, game theory, business studies, and behavioral theories, offering an interdisciplinary approach to analyzing business networks. Through simulations, we examine what happens when businesses with different tendencies to cooperate form networks, which is increasingly important for growing digital environments like digital platforms or the metaverse. Particularly, we observe how different levels of propensities to connect and cooperate lead to different network characteristics, including its structure, evolution, reciprocity, complexity, adaptation, symmetry, and stability. One of the most interesting findings was that low cooperation propensities lead to scale-free networks, where a few companies dominate—a reflection of the 80/20 rule commonly observed in economics and social structures. The 80/20 rule, also known as the Pareto Principle, suggests that 20% of participants (in this case, businesses) often generate 80% of the results or benefits, while the majority see much less success. On the other hand, companies that fail to form connections or build relationships with uncooperative partners in the early stages tend to fail, much like real-world observations where most businesses that fail do that within the first year. This type of uncooperative network mirrors those dynamics. By contrast, high cooperation levels lead to more cohesive and stable networks. This has practical implications for businesses: companies that focus on long-term, cooperative relationships not only tend to build stronger networks but also enhance their chances of survival and resilience in the face of change. In an era where digital platforms and the metaverse are reshaping how companies interact, these insights are more important than ever. Main takeaways: - Business Network Structure: Higher connection and cooperation levels create denser, more collaborative networks, while low cooperation often leads to unequal, scale-free networks dominated by a few actors. - Network Evolution: High connection and cooperation levels drive rapid network growth and stable, trust-based relationships. Lower relationship costs make it easier to maintain and expand connections. - Reciprocity: Increased cooperation fosters reciprocal, mutually beneficial relationships, promoting trust and equitable resource distribution. Low reciprocity can result in 80/20 outcome splits. - Complexity: Higher cooperation and connection levels, along with lower relationship costs, create more complex networks with a variety of relationships and interactions. - Adaptation: Cooperative networks adapt more easily to changes, with actors adjusting connections and dropping uncooperative partners as needed. - Cooperation and Conflict: High cooperation reduces conflicts and promotes a collaborative culture, while lower relationship costs help maintain positive relationships and minimize disputes. - Symmetry: High cooperation leads to more balanced, symmetric networks, whereas low cooperation results in asymmetric structures with a few dominant actors. - Convergence and Diversification: High connection and cooperation levels lead to convergence into clusters with shared goals, while low levels create more fragmented, diverse networks. - Stability: Networks with high cooperation are more stable, with fewer relationship terminations, while lower cooperation levels lead to instability and fluctuating network clusters.

Featured Image

Why is it important?

The theoretical findings directly translate into practical implications. As companies increasingly rely on digital platforms (which are a form of a network in itself), their success does not necessarily depend only on the products they offer but also on how well they manage relationships within their networks. This research shows that businesses must assess potential partners' willingness to cooperate, particularly in the early stages, as forming connections with uncooperative actors can lead to losses, much like real-world observations where many businesses fail within their first year. However, it is equally a deterrent to not forming relationships. Our findings also reveal the possibility of targeted interventions to enhance cooperation. Network managers can foster better collaboration by implementing incentives, trust-building initiatives, and relationship management strategies. For example, reducing administrative or transaction costs can encourage companies to maintain and expand their connections, promoting stronger, more cohesive networks. This is especially valuable for new ventures that may have experienced early setbacks from uncooperative partners. Additionally, businesses can actively shape the structure of their networks by facilitating strategic partnerships and forming clusters of cooperation. By promoting mutually beneficial interactions, rewarding collaboration, and building a culture of trust, companies can increase reciprocity, which in turn fosters stability and resilience within the network. The practical implications are clear: companies should focus on building long-term, cooperative relationships, not only to strengthen their own positions but also to encourage network growth and stability. Managing relationship costs, supporting cooperative behavior, and adapting to evolving network dynamics are key to maintaining successful, long-term partnerships. Businesses that adopt these strategies will find it easier to thrive in competitive, digitally connected environments like the metaverse. However, these insights are also important for network managers. The saying "We all do better when we all do better" applies directly here—when businesses in a network cooperate, the entire network becomes more resilient and prosperous. Network managers can play a pivotal role by promoting collaboration and reciprocity across all actors. Through targeted interventions like reducing barriers to cooperation, fostering trust-building activities, and encouraging strategic connections, they can cultivate a network where all members benefit. In essence, this research provides valuable insights and guidance for optimizing network strategies and helping businesses build resilient, high-performing networks. As digital platforms and the metaverse continue to reshape business interactions, these insights are more important than ever for companies seeking to foster growth, drive innovation, and navigate the complex landscape of interconnected business networks.

Perspectives

When we began this research, our primary goal was to create a model that better mimics human behavior and decision-making in business networks, particularly because of their relevance in an increasingly digitalized world. We wanted to provide a more realistic modeling approach to how businesses interact, including the ability to opt out of harmful relationships—a relatively novel feature introduced in game theory models and often overlooked in traditional approaches. We wanted to see what will happen with network characteristics when we include the opt-out strategy, heuristics, and imperfect updates. That is, we wanted to see the differences and similarities with the results of the (1) theoretical models that assume rationality and (2) results of empirical analyses of real-world networks. When we compare the network development in a round one hundred, we can see that the propensities to connect and to cooperate have a great impact on the network evolution and structure (see the figure below). Besides that, propensities to connect and cooperate shape network behavior and other characteristics, as well. Among them, we single out one here that we think is particularly important, both theoretically and practically: the prevalence of non-cooperation leads to a network structure following the 80/20 Pareto rule. It revealed that low cooperation results in a few dominant actors controlling most of the network’s success while the majority struggle—an outcome that mirrors many existing real-world economic and social patterns.

Katarina Kostelić

Read the Original

This page is a summary of: Modeling interactions in a dynamic heuristic business network, Applied Network Science, August 2024, Springer Science + Business Media,
DOI: 10.1007/s41109-024-00660-0.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page