The aim of the study was to validate a situational smoking typology by testing its efficacy in designing reduction strategies. Twenty-four volunteer smokers were scored on a Situational Smoking Questionnaire which allowed smokers to be classified into high- and low-activity groups on the basis of main cravings. High-activity smokers were further classified into those who smoked either under emotional stress or to aid concentration whilst low-activity smokers were subdivided into those who smoked to relieve boredom or to relax. From this situational model of motivation, alternative behavioural strategies tailored to smokers specific situational demands were devised that might achieve the same effect as smoking and so aid reduction. Smokers were then allocated at random to one of three treatment groups. 1. 1. A behavioural substitution group where alternative relaxation, emotional coping and distraction activities were rehearsed according to hierarchy of situational targets. 2. 2. A generalized coverant approach where emphasis was placed on generalized beliefs about smoking effects rather than situational ones. 3. 3. A no-treatment control group, where subjects only self-monitored their smoking over the treatment period. Smokers completed weekly diaries for baseline, four treatment sessions and 2- and 8-month follow-up periods. Both treatment groups showed a reduction in smoking levels over treatment sessions, but the behavioural group maintained a significantly lower level on session 4, and at follow-up. Those smokers initially classified as high-activity smokers benefited most from the behavioural programme. Measures of attitude towards smoking behaviour changed most significantly in the coverant group. The coverant approach appeared to increase motivation to cut down. and might in future studies be effectively employed together with the behavioural method. The results suggest generally that a differential situational model of smoking effects might be a more valid basis than the usual ‘fixed effect’ pharmacological model for the development of cessation programmes.