Follow the money: using payment behaviour as predictor for future self-exclusion

Behavioural tracking is an increasingly discussed method of early detection of potential gambling-related problems. Recent research has focused predominantly on information derived from gambling transactions and communication logs. Even though payment behaviour establishes a direct link to the financial consequences of gambling, it had thus far not become subject to this field of research. This article uses transactional data generated by 2696 customers from the online gambling label bwin.com to investigate how far specific payment behaviours and specific payment methods enable a prospective differentiation between self-excluders and controls. The number and amount of deposits, the variance of withdrawals, the amount of funds subject to reversed withdrawals and the usage of mobile phone billing were found to be positively associated with self-exclusion; the number of active gambling months and the usage of electronic wallets and prepaid cards negatively. Customer age displayed a U-shaped relation, with customers of below and above average being stronger associated with self-exclusion. However, it must be noted that the validity of the resulting predictive model derived from payment data is lower than comparable models derived from gambling data. The results are discussed in the light of implementing early-detection and intervention algorithms based on customer behaviour.