AIMS/HYPOTHESIS: Women with diabetes remain at increased risk of adverse pregnancy outcomes associated with poor pregnancy preparation. However, women with type 2 diabetes are less aware of and less likely to access pre-pregnancy care (PPC) compared with women with type 1 diabetes. We developed and evaluated a community-based PPC programme with the aim of improving pregnancy preparation in all women with pregestational diabetes. METHODS: This was a prospective cohort study comparing pregnancy preparation measures before and during/after the PPC intervention in women with pre-existing diabetes from 1 June 2013 to 28 February 2017. The setting was 422 primary care practices and ten National Health Service specialist antenatal diabetes clinics. A multifaceted approach was taken to engage women with diabetes and community healthcare teams. This included identifying and sending PPC information leaflets to all eligible women, electronic preconception care templates, online education modules and resources, and regional meetings and educational events. Key outcomes were preconception folic acid supplementation, maternal HbA1c level, use of potentially harmful medications at conception and gestational age at first presentation, before and during/after the PPC programme. RESULTS: A total of 306 (73%) primary care practices actively participated in the PPC programme. Primary care databases were used to identify 5075 women with diabetes aged 18-45 years. PPC leaflets were provided to 4558 (89.8%) eligible women. There were 842 consecutive pregnancies in women with diabetes: 502 before and 340 during/after the PPC intervention. During/after the PPC intervention, pregnant women with type 2 diabetes were more likely to achieve target HbA1c levels ≤48 mmol/mol (6.5%) (44.4% of women before vs 58.5% of women during/after PPC intervention; p = 0.016) and to take 5 mg folic acid daily (23.5% and 41.8%; p = 0.001). There was an almost threefold improvement in 'optimal' pregnancy preparation in women with type 2 diabetes (5.8% and 15.1%; p = 0.021). Women with type 1 diabetes presented for earlier antenatal care during/after PPC (54.0% vs 67.3% before 8 weeks' gestation; p = 0.003) with no other changes. CONCLUSIONS/INTERPRETATION: A pragmatic community-based PPC programme was associated with clinically relevant improvements in pregnancy preparation in women with type 2 diabetes. To our knowledge, this is the first community-based PPC intervention to improve pregnancy preparation for women with type 2 diabetes. DATA AVAILABILITY: Further details of the data collection methodology, individual clinic data and the full audit reports for healthcare professionals and service users are available from https://digital.nhs.uk/data-and-information/clinical-audits-and-registries/our-clinical-audits-and-registries/national-pregnancy-in-diabetes-audit .