The quality of primary data sources is a critical prerequisite for evaluating cycling interventions, modelling cycling behaviour, and supporting evidence-based urban mobility planning in pursuit of sustainable transport policies. This study examines the concept of representativeness in cycling-related data by comparing four key data sources from the same year: a household travel survey, a cyclist`s survey, automatic counting stations, and GPS tracking data. A methodological framework is developed to assess data quality and is applied to empirical datasets from the city of Dresden, Germany. The findings reveal that no single source offers a universally comprehensive view; instead, the suitability of each dataset varies according to the planning objective. GPS-based data provides high spatial and temporal resolution, making it especially valuable for network design and traffic analysis, while household surveys and cyclists surveys offer essential socio-demographic insights and, in some cases, subjective user experiences. Automatic counters lack spatial representativeness but provide valuable insights into daily, weekly, and seasonal variations in cycling volumes. The study highlights the value of mixed-method approaches to overcome the limitations of individual data types and to ensure a more holistic understanding of cycling behaviour. Future research should prioritise improving the demographic representativeness of digital data and advancing validation techniques to strengthen the reliability and utility of cycling data for planning and policy.