Surf tourism is of major importance to the tourism industry. Nevertheless, very few investigations of the surf tourism market exist. This paper extends the work by Fluker (2003) and Dolnicar and Fluker (2003) by investigating surf tourists from a behavioural perspective with the main aim of the study being to gain an insight into the travel patterns of the surf tourism market. This is achieved in an empirical way by using unsupervised neural networks to partition a group of surfers into homogeneous segments based on their past surf destination choice. This binary information was gathered by means of an online survey, which asked respondents questions indicating whether or not they have ever surfed in particular places. In addition, descriptive information is included in the data set and is divided into “surf related questions”, “personal characteristics” and “travel behaviour”. It was found that based on past destination choice, six market segments could be described, each with significantly different ages, surfing ability, length of stay, preferred wave type, and regularity of undertaking surf trips. The results of these finding have implications for both surf destinations and the tourism industry that facilitates the experience.