OBJECTIVES Computerised cognitive behaviour therapy (cCBT) forms a core component of stepped psychological care within primary care in the UK and other countries. However, the existing clinical effectiveness evidence for cCBT comes from developer-led trials and independent research is needed which evaluates the clinical and cost-effectiveness of cCBT. METHODS A cost-effectiveness analysis was undertaken comparing two cCBT software packages, free-to-use MoodGYM and a commercial pay-to-use Beating the Blues (BtB), in addition to usual general practitioner care (UGPC), with UGPC alone, for the treatment of depressed adults. The analysis was based on data collected on the Randomised Evaluation of the Effectiveness and Acceptability of Computerised Therapy (REEACT). REEACT was a large (n=691), pragmatic multicentre study, independently conducted in a primary care setting.  Outcomes were assessed using EQ-5D and used to estimate quality-adjusted life-years (QALYs). Resource use and costs were estimated from a NHS and Personal Social Services perspective. Scenario analyses were performed to determine the impact on cost-effectiveness of alternative assumptions, and uncertainty was characterised using cost-effectiveness acceptability curves. RESULTS BtB was both more expensive and generated lower QALYs than UGPC alone (dominated) and MoodGYM yielded lower QALYs but at lower cost, resulting in an ICER of £6,933 per additional QALY for UGPC alone versus MoodGYM.  UGPC alone was the most cost-effective intervention in the majority of scenario analyses, and the intervention most likely to be cost-effective at a £20,000 per QALY threshold (probabilities ranging across scenarios from 0.545 to 0.619).  CONCLUSIONS Our findings indicate that commercially-produced products were no more effective than free-to-use cCBT programmes.  Importantly, neither BtB or MoodGYM appeared cost-effective compared to UGPC alone. Practice recommendations such as those offered by NICE and other countries supporting the use of cCBT within stepped models of care for depression will need to be reconsidered in light of these results.