Organizations increasingly rely on digital data and algorithmic technologies to control workers. Most empirical accounts of this phenomenon focus on digital platforms, contexts where digital data such as geolocation and customer ratings effectively ‘stand in’ for reality. As a result, less is known about how digital control unfolds in contexts that require workers to engage in translation practices - situated and materially mediated efforts to represent physical labor into digital data. To investigate this issue, we conducted a 24-month ethnography of a logistics organization in the truck transportation industry. In this setting, workers such as truck drivers, dispatchers, material handlers, and customer service agents continuously engage in translation practices to ensure that flows of merchandise across space and time remain successfully aligned with digital representations. We found that actors engaged in two sets of translation practices: digitizing the physical, where physical phenomena were translated into digital data through algorithmic manipulations, and physicalizing the digital, where physical phenomena were reconstructed from digital data, often using photos and text. These practices enabled the digitization of physical phenomena across increasing spatial and temporal scales: actions, events, and work processes. In the realm of actions, control was brittle: hard in coercive measures yet prone to breakdowns. Conversely, as the scale of physical phenomena grew, control became more ductile: less coercive yet harder to break. This is because the continuous involvement in translation practices sustained the internalization of two normative expectations: a ‘continuous improvement’ mindset and constant accountability. By explaining the role of translation practices and the importance of scale in shaping digital control, this study provides two important theoretical contributions to the literature on digital control.
