Data Access Control in Personal Data Ecosystems: A Business Model Perspective

The transition towards Personal Data Ecosystems (PDEs) requires sustainable business models that balance data sovereignty between data subjects and providers. This research applies a business model perspective to data sovereignty and PDE research, offering a comprehensive framework for understanding the strategic decisions data providers make regarding data access control in PDEs. Concretely, we investigate the business dimensions that influence data providers' willingness to grant data access control to data subjects via a two-staged methodology. In a first step, 25 interviews identified key business dimensions, representing a trade-off between value proposition (user- and ecosystem value), value network (collaboration and competition), value finance (value capturing and privacy risk) and value architecture (coreness of data, level of processing of data). In a second stage, a use case analysis of the Personal Data Store (PDS) in a mobility PDE was performed, utilizing an Analytic Hierarchy Process (AHP) to quantify the preferences of data providers within a mobility ecosystem involving 21 mobility and data experts. This data shows the findings of the AHP analysis, which show value proposition and value finance are the most salient dimensions in this mobility ecosystem.

The analytical hierarchical process (AHP) methodology was employed to research the preferences of data providers in the mobility ecosystem. In this work, the AHP was used to determine the preference of the business dimensions for granting data access control . The preferences of data providers were ranked by experts through a pairwise comparison .Next, respondents assessed the relative importance or preference using a numerical scale ranging from 1 to 9. The AHP was conducted throughout telephone interviews with 21 mobility and data experts in Belgium and the Netherlands. These pairwise comparisons are transformed into individual result matrices, which are used to calculate leading eigen vectors to determine the relative weights of the dimensions. The analysis was performed on a group level, avoiding bias that may be present when the judgements are considered from a single expert. The findings were examined within the overall group that encompassed all participants and within the MAAS and C-ITS subgroups. Using the geometric mean of the individual result matrices, the group matrices are calculated of the different stakeholder groups. The consistency ratio was calculated, for which 0.1 is considered to indicate a tolerable consistent ranking for grouped responses . The “ahpsurvey” package in R was used to analyze the results.