A proportion of gamblers experience problems. The role of overconsumption in developing gambling problems is sparsely described in the literature and there is little scientific knowledge about the prevention of gambling problems. There are some promising results regarding personalized feedback on gambling habits, and there is a need for more research. The overall aim of this thesis was to explore the role of overconsumption in problem gambling and target it in a preventive intervention. The preventive intervention was to give gambling consumption feedback to high consumers in order to make them reflect upon their gambling habits and enhance their motivation for change. Study I aimed to explore the dimensionality of GamTest, an online test of gambling behaviour, and validate it against PGSI and the gambler’s own perceived problems. Data came from four Nordic gambling sites, n = 10,402. In an ESEM analyses, GamTest had a high degree of correspondence with the players’ own perceived problems and with the PGSI. In an EFA, GamTest captured five dimensions of problematic gambling (i.e. overconsumption of money and time, and negative financial, social and emotional consequences). A bifactor approach showed a general factor and four specific residual factors, negative emotional consequences contribute to the dominant part of the general factor. Study II aimed to examine both the psychometric properties of the Jonsson-Abbot Scale (JAS) and its predictive validity with respect to increased gambling risk and problem gambling onset. The results are based on repeated interviews with 3,818 participants within the Swedish longitudinal gambling study. The results indicate an acceptable fit of a three-factor solution in a CFA, with ‘Overconsumption (OC),’ ‘Gambling fallacies (GF),’ and ‘Reinforcers (RI)’ as factors. When controlled for risk potential measured at baseline, GF and RI were significant predictors of gambling risk potential, and GF and OC were significant predictors of problem gambling onset at 12-month follow up. Study III’s primary objective was to investigate the effects of providing personalized feedback on gambling intensity among high consumers in Norway. An RCT design was used to evaluate how behavioural feedback by telephone or letters affects subsequent gambling expenditure. A sample of 1,003 statistical matched triplets, from the top 0.5 % of customers, were randomly assigned to telephone, letter, or a no-contact control condition. Over 12 weeks, theoretical loss decreased 29 % for the telephone, and 15 % for the letter, conditions, compared with 3 % for the control group. Study IV was a 12-month follow-up of Study III, aimed to investigate the relative effects over twelve months. The telephone group showed a 30 % reduction in theoretical loss, the letter group 13 %, both outperforming the control group with a 7 % reduction. Less than 1% in all groups stopped playing at Norsk Tipping. These four studies indicate that overconsumption of gambling plays different roles in problem gambling. The role of overconsumption in preventing gambling problems is discussed. Contacting high consumers about their gambling expenditure appears to be an effective method for gambling companies to meet their duty of care for customers. Technical evolution has made it possible for gambling companies to fulfil their duty of care, but this has to be regulated and mandatory if it is to be effective.