There are few effective interventions for dementia. To determine the clinical effectiveness and cost-effectiveness of an intervention to promote self-management, independence and self-efficacy in people with early-stage dementia. To undertake a randomised controlled trial of the Journeying through Dementia intervention compared with usual care, conduct an internal pilot testing feasibility, assess intervention delivery fidelity and undertake a qualitative exploration of participants' experiences. A pragmatic two-arm individually randomised trial analysed by intention to treat. A total of 480 people diagnosed with mild dementia, with capacity to make informed decisions, living in the community and not participating in other studies, and 350 supporters whom they identified, from 13 locations in England, took part. Those randomised to the Journeying through Dementia intervention (n = 241) were invited to take part in 12 weekly facilitated groups and four one-to-one sessions delivered in the community by secondary care staff, in addition to their usual care. The control group (n = 239) received usual care. Usual care included drug treatment, needs assessment and referral to appropriate services. Usual care at each site was recorded. The primary outcome was Dementia-Related Quality of Life score at 8 months post randomisation, with higher scores representing higher quality of life. Secondary outcomes included resource use, psychological well-being, self-management, instrumental activities of daily living and health-related quality of life. Participants were randomised in a 1 : 1 ratio. Staff conducting outcome assessments were blinded. Outcome measures were administered in participants' homes at baseline and at 8 and 12 months post randomisation. Interviews were conducted with participants, participating carers and interventionalists. The mean Dementia-Related Quality of Life score at 8 months was 93.3 (standard deviation 13.0) in the intervention arm (n = 191) and 91.9 (standard deviation 14.6) in the control arm (n = 197), with a difference in means of 0.9 (95% confidence interval -1.2 to 3.0; p = 0.380) after adjustment for covariates. This effect size (0.9) was less than the 4 points defined as clinically meaningful. For other outcomes, a difference was found only for Diener's Flourishing Scale (adjusted mean difference 1.2, 95% confidence interval 0.1 to 2.3), in favour of the intervention (i.e. in a positive direction). The Journeying through Dementia intervention cost £608 more than usual care (95% confidence interval £105 to £1179) and had negligible difference in quality-adjusted life-years (-0.003, 95% confidence interval -0.044 to 0.038). Therefore, the Journeying through Dementia intervention had a mean incremental cost per quality-adjusted life-year of -£202,857 (95% confidence interval -£534,733 to £483,739); however, there is considerable uncertainty around this. Assessed fidelity was good. Interviewed participants described receiving some benefit and a minority benefited greatly. However, negative aspects were also raised by a minority. Seventeen per cent of participants in the intervention arm and 15% of participants in the control arm experienced at least one serious adverse event. None of the serious adverse events were classified as related to the intervention. Study limitations include recruitment of an active population, delivery challenges and limitations of existing outcome measures. The Journeying through Dementia programme is not clinically effective, is unlikely to be cost-effective and cannot be recommended in its existing format. Research should focus on the creation of new outcome measures to assess well-being in dementia and on using elements of the intervention, such as enabling enactment in the community. This trial is registered as ISRCTN17993825. This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 26, No. 24. See the NIHR Journals Library website for further project information.