ResearchPod Summary
Business models are complex systems of value creation, delivery, and capture. In a digital context, these models are increasingly defined by how firms interact with demand-side data. Rather than viewing markets as simple equilibrium points, this perspective emphasizes that markets must be actively discovered, sustained, and shaped. Digital infrastructure—such as cloud computing and real-time analytics—serves as the backbone for these interactions, allowing firms to move beyond static offerings toward dynamic, data-driven service ecosystems.
Data is not merely extracted; it is co-created within the customer relationship. The paper categorizes the data journey into three stages: tokens (raw events), objects (structured metrics), and commodities (service-ready products). A critical managerial decision is determining the level of control to retain at each stage. Firms must navigate a 2x2 matrix based on the source of data (software vs. physical machines) and the required reaction speed (reflection vs. real-time action). This governance choice dictates whether a firm operates as a passive analyst or an active controller of customer behavior.
Modern consumer behavior is increasingly characterized by Chronic Consumer Liminality, where individuals experience frequent, non-linear life transitions that destabilize traditional routines. This shift toward liquid consumption creates opportunities for firms to offer personalized, hyper-local services. However, this trend is inextricably linked to surveillance capitalism, where private human experience is claimed as raw material for behavioral data. Firms are shifting from monitoring to 'actuating'—using subtle cues and rewards to herd consumer behavior toward profitable outcomes.
[[RP_SECTION:data-as-a-refinery|Data as a refinery]]
Sam: The traditional view of data as a passive resource—something to be extracted like oil—is wrong in a specific, consequential way. A recent framework on digital business models makes the case that the customer isn't the mine. The customer is the refinery.
Alex: That's a sharp reframe. What does it actually change about how a firm governs its strategy?
Sam: It shifts the governance challenge from extraction to design. If you treat data like oil, you assume it exists independently of your actions, waiting to be mined. But in modern digital ecosystems, data is co-created with the customer. You aren't just measuring behavior—you are designing the digital architecture that produces the behavior you then measure. The firm isn't observing the market. It's shaping the conditions that generate the signal.
Alex: Which means the firm has a much more active role than the standard "data-driven strategy" framing implies.
Sam: Exactly. The framework calls this the transition from monitoring to actuating. You move beyond prediction to actively tuning and conditioning user behavior through what it terms economies of action. The firm isn't just responding to demand—it's engineering the conditions that produce it. That's a meaningful shift in the power dynamic between firm and customer, and it changes what "strategy" even means at the architectural level. [[RP_SECTION:governance-and-temporal-constraints|Governance and temporal constraints]]
Alex: If I'm mapping a company's strategy against this, the framework offers a two-by-two based on software versus machines and reflection versus real-time. How does that diagnostic actually help?
Sam: It's a tool to prevent the one-size-fits-all fallacy. You have to ask two questions: where is the data created, and what reaction speed does the system require? Pure software contexts can often operate on a reflective cycle—hours or weeks of analysis. But if your data comes from physical sensors, the system may demand real-time actuation in milliseconds. Those are structurally different governance problems, and conflating them is where firms go wrong.
Alex: So the governance model has to match the temporal constraints of the underlying system.
Sam: Precisely. If you apply a slow, reflective governance model to a system that requires millisecond-level sensor actuation, you're not just inefficient—you're misaligned at the architectural level. Take a smart-appliance manufacturer. They might start by collecting data for product improvement, which is reflective. But once they shift to real-time sensor telemetry, they gain the ability to adjust energy consumption during peak hours automatically. At that point, the right question is: is this value creation for the user, or is the firm herding the user to optimize its own infrastructure costs? The data loop looks identical from the outside. The intent is what distinguishes them. [[RP_SECTION:chronic-consumer-liminality|Chronic consumer liminality]]
For managers and researchers, the core challenge is balancing the lucrative potential of data-driven personalization with the ethical risks of eroding consumer trust. As consumers become more privacy-conscious, business models that rely on the exploitation of behavioral data face long-term viability risks. Encouraging critical thinking and moral judgment in business design is essential to avoid market failure and ensure that digital services contribute positively to society.
AI-generated third-party summary by ResearchPod. Not official content or an endorsement by the paper authors or affiliated organizations.
Alex: That forces you to look past the technical architecture and interrogate the purpose behind the design. How does this connect to the concept of Chronic Consumer Liminality?
Sam: Chronic Consumer Liminality describes a state where consumers are in constant, non-linear transition—routines are fragmented, long-term preferences are less stable, and people are more responsive to immediate cues. When your behavioral anchors are disrupted, you become more susceptible to the nudging that actuating systems provide. The framework argues you become more liquid—more easily conditioned by the architecture around you.
Alex: That's a sobering implication. The more we live in that state of transition, the more we effectively cede behavioral agency to automated systems. Is there any structural counterforce, or does the framework treat actuating as the inevitable endpoint?
Sam: There is a counterforce noted, though it's speculative at this stage. As consumers become more aware of surveillance capitalism, they become more selective about the permissions they grant. The argument is that we may see a market segment where un-actuated experiences become a premium product—firms competing on data sovereignty, explicitly refusing to herd or condition behavior, with privacy positioned as a differentiator rather than a compliance cost. [[RP_SECTION:competitive-moats-and-infrastructure|Competitive moats and infrastructure]]
Alex: That's a compelling angle. But as a researcher I'd want to push on the framework's own limits. There's a conflation here between digital infrastructure—the basic plumbing, standard analytics, commodity tooling—and enabling technology, meaning the proprietary value driver. Without a cleaner taxonomy there, how do you distinguish a defensible competitive moat from a cost-of-entry utility?
Sam: That's where a careful referee would push back, and the framework doesn't fully resolve it. It doesn't isolate what constitutes a genuine moat versus what is foundational plumbing that every competitor eventually acquires. The risk is that a firm—or an analyst—labels commodity infrastructure as the source of competitive advantage when it's actually just table stakes. The real value, if the framework is right, lies in the specific, proprietary way a firm governs its data loops: the design choices, the feedback architecture, the behavioral conditioning that is hard to replicate. But the framework doesn't give you a clean test for when you've crossed that line.
Alex: So the practical upshot for a researcher is to stop treating data as a static asset and start auditing the governance design itself—mapping the firm's data strategy against its operational reality to determine whether the loop is serving the user or engineering the user.
Sam: That's the core of it. Once you can map that, you can see the business model for what it is: an engineering project for human behavior. The data is just the feedback signal. The architecture is the strategy. And the real analytical work lies in distinguishing firms with genuinely proprietary architectures from those running on shared plumbing—because those two things can look identical until the competitive pressure arrives.
Alex: Thanks for listening to ResearchPod.