Our article, “Evaluation is key: a survey on evaluation measures for synthetic time series,” accepted for publication in the International Journal of Big Data, aims to clarify the evaluation of synthetic data generation for time series. Synthetic data generation models the distributions of real datasets to create new data, which is crucial in privacy-sensitive fields like healthcare. While image synthesis has been extensively studied, time series synthesis is equally vital for practical applications. Despite the availability of numerous models and measures, there is no consensus on defining or quantifying high-quality synthetic time series. Our comprehensive survey reviews various evaluation measures, provides clear definitions, organizes them into a taxonomy, and offers guidance on selecting appropriate measures.