The internet is often understood as a tool that allows users to access information, communicate, and perform various tasks. While this description may have been accurate in its earlier stages, it no longer captures the nature of the contemporary digital environment. The internet has undergone a structural transformation. It is no longer a passive system that simply responds to user input. Instead, it has become an active environment that shapes how users interact, make decisions, and behave.
This transformation is largely driven by the rise of digital platforms. These platforms do not merely host content. They organise, filter, and prioritise information through algorithmic systems designed to maximise user engagement. As a result, the internet now operates as a system in which user behaviour is continuously observed, measured, and optimised.
This essay argues that the modern internet functions as an economy of behaviour. In this system, user actions are not incidental but central. Behaviour is collected as data, analysed to identify patterns, and used to influence future interactions. Through continuous feedback loops between users and algorithmic systems, platforms do not simply respond to behaviour. They actively shape it. This shift raises important questions about autonomy, influence, and the governance of digital environments.
From Infrastructure to Platform Systems
In its early form, the internet functioned as a decentralised network that enabled access to information. Users interacted with relatively static webpages, and the flow of information was largely one-directional. Content existed independently of individual users, and interactions were limited in scope.
The development of platform-based systems has fundamentally altered this structure. Today, much of online activity takes place within platforms that mediate how information is accessed and experienced. These platforms act as intermediaries, determining what content is visible and how it is presented to users.
Unlike earlier versions of the internet, platforms rely on algorithmic curation. Content is not displayed in a neutral or chronological order. Instead, it is ranked based on predicted relevance and the likelihood of user engagement. These predictions are generated through the analysis of behavioural data, including past interactions, viewing patterns, and inferred interests.
As a result, users no longer navigate an open information space. They interact with environments that are continuously shaped by data-driven systems. Visibility is no longer determined by what exists, but by what the system selects.
This shift marks a transition from an information-based system to a platform-dominated environment in which access to information is structured and controlled.
Behaviour as an Economic Resource
At the centre of platform systems is the transformation of user behaviour into an economic resource. Platforms generate value not only through the content they host but through the interactions users have with that content.
Every user action, including clicking, liking, sharing, pausing, or scrolling, produces data. These behavioural signals are collected and analysed to identify patterns. Unlike traditional forms of data, which rely on explicit inputs such as stated preferences or demographic information, behavioural data is implicit, continuous, and often more revealing. It reflects what users actually do rather than what they claim to want.
This distinction is significant. In traditional economic systems, value is often derived from declared preferences. In platform-based systems, however, value is derived from observed behaviour. Users may express certain interests, but their real engagement patterns often differ. Platforms prioritise behavioural signals because they provide a more accurate and immediate representation of attention.
Behavioural data is particularly valuable because it captures engagement in real time. It reflects hesitation, repetition, curiosity, and distraction. These signals allow platforms to detect patterns that are inconsistent, impulsive, and often subconscious. As a result, platforms do not need to fully understand users in terms of identity or intention. Instead, they rely on patterns of interaction that can be measured, modelled, and predicted.
Once behaviour becomes predictable, it becomes economically valuable. Platforms operate within markets where advertisers pay for access to user attention. The more precisely a platform can predict and influence behaviour, the more effectively it can target advertisements and maximise revenue.
In this context, behaviour functions as a form of currency. It is continuously extracted, analysed, and reinvested into the system to improve its predictive capabilities. This transformation marks a shift from an information-based economy to one centred on behavioural data.
The Feedback Loop of Behavioural Optimisation
Platform systems operate through a continuous feedback loop between users and algorithmic processes.
When a user interacts with content, that interaction is recorded and used to update the system’s model of the user. The system then adjusts the content that is presented, prioritising items that are more likely to generate engagement. The user responds to this updated environment, producing further data, and the cycle continues.
This process is often described as personalisation, but it is more accurately understood as behavioural optimisation. The system is not simply matching content to user preferences. It is continuously refining the environment to maximise engagement outcomes.
As the system accumulates more data, its predictions become increasingly precise. It learns not only what types of content a user engages with, but under what conditions engagement is most likely to occur. This includes factors such as timing, format, and context. Over time, the system becomes more effective at anticipating user behaviour before it happens.
This does not merely improve relevance. It alters the structure of the environment itself. Content that generates engagement is reinforced through increased visibility, while content that does not is gradually deprioritised. As a result, users are exposed to patterns that become more stable and predictable over time.
Importantly, this feedback loop is self-reinforcing. Past behaviour is used to justify future exposure, which in turn produces similar behaviour. This creates a cycle in which users are continuously guided toward certain types of interaction without explicit direction.
The process also becomes increasingly invisible. Users experience the system as intuitive or personalised, rather than as a structured environment. This obscures the extent to which behaviour is being shaped rather than simply expressed.
The Structuring of Choice
Although users continue to make choices within digital environments, those choices are structured by the system.
Before a user selects content, the system has already filtered and ranked the available options. This determines what is visible and what is not. As a result, user decisions are made within a curated set of possibilities.
Because this curation is based on behavioural data, it often appears personalised. Content seems relevant and aligned with user interests. However, this relevance is the result of algorithmic selection rather than independent exploration.
The structuring of choice does not remove autonomy, but it shapes it. Users are more likely to engage with content that is prominently displayed and less likely to encounter content that is not prioritised.
Over time, this can reinforce existing patterns of behaviour. Users are repeatedly exposed to similar types of content, which can narrow the range of perspectives and experiences they encounter.
Case Study: Instagram and Algorithmic Curation
Instagram provides a clear example of how behavioural optimisation operates in practice.
The platform does not display content in chronological order. Instead, it ranks posts based on a range of signals, including past interactions, time spent viewing content, and inferred interests. Features such as the Explore page and Reels feed are specifically designed to surface content that is most likely to capture user attention.
Consider a user who begins engaging with fitness-related content. Initially, the content may be broad, including general workout routines and health advice. However, as the system collects data on the user’s interactions, it begins to prioritise content that generates higher levels of engagement.
Over time, this leads to a narrowing of content. Posts become more specific, more visually refined, and often more extreme in their representation. The system does not intentionally direct users toward a particular outcome. Instead, it amplifies content that performs well within the engagement-based ranking system.
This process can have cumulative effects. Repeated exposure to similar types of content can shape users’ perceptions of what is typical or desirable. Content that initially appears aspirational may begin to feel normal through repetition. This occurs not through direct persuasion, but through gradual shifts in visibility and familiarity.
The result is a form of indirect influence. Users are not explicitly told what to think or do. Instead, they are immersed in environments where certain ideas, behaviours, and standards are more prominent than others.
The case of Instagram illustrates how platforms do not simply reflect user behaviour. They actively shape it by structuring the conditions under which behaviour occurs.
Alignment of Incentives and Emergent Outcomes
The optimisation of engagement creates a structural alignment between platform incentives and certain types of content.
Content that generates strong emotional reactions tends to perform better. It is more likely to be shared, commented on, and engaged with. As a result, platforms are more likely to prioritise such content.
This does not require intentional bias. It is an outcome of how the system is designed to function.
However, this alignment can produce unintended consequences. For example, misleading or extreme content may spread more rapidly because it captures attention more effectively. Similarly, polarising content may become more visible than balanced or nuanced perspectives.
These outcomes are not necessarily the result of deliberate decisions. They emerge from the interaction between user behaviour and algorithmic optimisation.
This highlights a key tension. The objective of maximising engagement does not always align with broader social goals such as accuracy, diversity of information, or user well-being.
Governance Challenges
The transformation of the internet into a behavioural system presents significant challenges for governance.
Platforms exercise influence by shaping what users see and how they interact. However, they are not governed in the same way as traditional institutions. Their systems are complex, dynamic, and often opaque.
Regulatory frameworks are typically designed to address clear actions and responsibilities. Platform systems, by contrast, produce outcomes through interactions between algorithms and millions of users.
This makes it difficult to assign responsibility. When harmful outcomes occur, it is not always clear whether they result from individual actions, system design, or emergent behaviour.
Additionally, platforms operate across national boundaries, making regulation more complex. Their systems evolve rapidly, often outpacing the development of policies and regulations.
Efforts to address these challenges include proposals for increased transparency, accountability, and oversight. However, these approaches face limitations when applied to systems that continuously adapt and learn.
Conclusion
The internet has evolved from a tool for accessing information into a system that shapes behaviour.
Platforms collect and analyse user interactions, using this data to predict and influence future activity. Behaviour is no longer incidental. It is central to how value is created within digital environments.
Through continuous feedback loops, users and platforms influence each other. While users shape the system through their actions, the system simultaneously shapes user behaviour through its structure and recommendations.
Understanding the internet as an economy of behaviour provides a clearer framework for analysing its impact. It highlights how digital environments are designed and how they influence user experience.
As these systems continue to develop, it becomes increasingly important to consider how they should be governed. The challenge lies not only in understanding how they operate but also in ensuring that their design aligns with broader social values.
References (APA 7)
References
Bennett, W. L., & Livingston, S. (2018). The disinformation order: Disruptive communication and the decline of democratic institutions. European Journal of Communication, 33(2), 122–139. https://doi.org/10.1177/0267323118760317
Flew, T. (2021). Regulating platforms. Polity Press.
Gillespie, T. (2018). Custodians of the internet: Platforms, content moderation, and the hidden decisions that shape social media. Yale University Press.
Nissenbaum, H. (2010). Privacy in context: Technology, policy, and the integrity of social life. Stanford University Press.
Solove, D. J. (2006). A taxonomy of privacy. University of Pennsylvania Law Review, 154(3), 477–564.
Van Dijck, J., Poell, T., & de Waal, M. (2018). The platform society: Public values in a connective world. Oxford University Press.
Be the first to comment