Background Longitudinal research has many strengths, but is often compromised by attrition of participants. This is a particular problem in studies with 'difficult-to-reach' populations. Recommended techniques for maximizing retention are often impractical for routine clinical services. Aims To examine whether methods that are available in routine practice can achieve satisfactory follow-up of a cohort of patients, 11 months after assessment for alcohol treatment. Methods Using a longitudinal cohort design, we tested the incremental effect of a simple three-step approach with 124 consecutive patients seeking treatment for alcohol problems: first an interview appointment, then postal questionnaires and, finally, a reminder by telephone or letter. Data on treatment compliance and self-reports of drinking behaviour during the 11-month follow-up period were obtained. Results Stage of contact was significantly associated with treatment compliance and treatment outcome. The final telephone reminder proved to be an important step in obtaining a sample that more accurately reflected both the treatment attendance and outcome of the original cohort. Conclusions Intensive resources, contact with multiple agencies and financial incentives were not necessary to achieve a contact rate of 75% and successful follow-up of 67%. Alcohol treatment outcome research is feasible within services with limited time and resources.