There is no meta-analysis of cognitive behavioural therapy for delusions and hallucinations separately. The aim of this meta-analysis is to evaluate the end-of-treatment effects of individually tailored case-formulation cognitive behavioural therapy on delusions and auditory hallucinations using symptom-specific outcome measures. A systematic search of the trial literature was conducted in MEDLINE, PSYCHINFO and EMBASE. Eighteen studies were selected with symptom specific outcome measures. Hedges' g was computed and outcomes were pooled meta-analytically using the random-effects model. Our main analyses were with the selected studies with CBT using individually tailored case-formulation that aimed to reduce hallucinations and delusions. The statistically significant effect-sizes were 0.36 with delusions and 0.44 with hallucinations, which are modest and in line with other recent meta-analyses. Contrasted with active treatment, CBT for delusions lost statistical significance (0.33), but the effect-size for CBT for hallucinations increased (0.49). Blinded studies reduced effect-size in delusions (0.24) and gained some in hallucinations (0.46). There was no heterogeneity in hallucinations and moderate heterogeneity in delusion trials. We conclude that CBT is effective in treating auditory hallucinations. CBT for delusions is also effective, but the results must be interpreted with caution, because of heterogeneity and the non-significant effect-sizes when compared with active treatment. © 2014 Elsevier B.V.