This study investigates how referral to Child and Adolescent Mental Health Services (CAMHS), and subsequent engagement relate to educational outcomes among young people in Wales. Using the Secure Anonymised Information Linkage (SAIL) Databank, we constructed a national cohort of individuals, enumerated in the 2011 Census, who received their first CAMHS referral between 2010 and 2020 at age 5–18. A matched general-population cohort was created using age, sex, deprivation, and locality. Linkage to the ONS 2011 Census Wales (CENW) person and household datasets via ALF and RALF provides detailed socioeconomic and demographic information, including household composition, parental employment, ethnicity, deprivation, free school meal eligibility, and urban/rural residence. Educational outcomes include academic attainment, fixed-term exclusions, permanent exclusion, and Pupil Referral Unit (PRU) attendance. Schooling follows fixed stages, thus, the timing of referral is essential. Children referred at younger ages contribute several years of post-referral data, enabling assessment of how early mental health difficulties influence later educational performance. Adolescents referred at ages 15–18 have limited or no post-referral schooling, their outcomes largely reflect pre-referral difficulties and factors that may precede service contact. Diagnostic categories, referral source, service contacts, and non-attendance (DNAs) are summarised to describe variation in pathways across age groups. This study provides new population-level evidence on the relationship between CAMHS referral patterns and school outcomes, accounting for developmental timing and social context. Findings demonstrate how linked administrative data can improve understanding of educational risks among young people with mental health needs, and appropriately inform more coordinated policy responses.