Privacy is a key aspect for the European Union (EU), where it is regulated by a specific law, the General Data Protection Regulation (GDPR). Compliance to the GDPR is a problem for organizations, it imposes strict constraints whenever they deal with personal data and, in case of infringement, it specifies severe consequences such as legal and monetary penalties. Such organizations frequently are complex systems, where personal data is processed by humans and technical services. Therefore, it becomes fundamental to consider privacy from the social perspective when designing such system, i.e., when relations between different components are specified. This is, indeed, also specified in the GDPR, which encourages to apply privacy-by-design principles. This paper proposes a method to support the design of GDPR compliant systems, based on a socio-technical approach composed of a modeling language and a reasoning framework.