Distributed Denial of Service (DDoS) attacks mean millions in revenue losses to many industries, such e-commerce and online financial services. The amount of reported DDoS attacks has increased with 47% compared to 2013. One of the reasons for this increase is the availability and ease of accessibility to websites, which provide DDoS attacks as a paid service, called Booters. Although there are hundreds of Booters available, current researches are focused on a handful sample of them - either to analyse attack traffic or hacked databases. Towards a thorough understanding and mitigation of Booters, a comprehensive list of them is needed. In this paper we characterize Booter websites and demonstrate that the found main characteristics can be used to classify Booters with 85% of accuracy. The Dutch National Research and Education Network (SURFnet) has been using a list generated by our methodology since 2013, what demonstrates high relevance to the network management community and the security specialists.