Motor proficiency and socioeconomic status are considered critical role players in setting children on an early trajectory of academic failure or success. To determine longitudinal influences of motor proficiency (MP) and socioeconomic status (SES) on academic achievement of primary school learners, a mediating model was constructed and tested with structural equation modelling (SEM). Primary school learners (N = 381) took part in this 7 school year NW-CHILD longitudinal study, with baseline measurements in grade 1 and follow-up measurements in grades 4 and 7. Academic achievement was based on school, national, and provincial assessments, and the "Bruininks Oseretsky Test of Motor Proficiency-2" ("BOT-2") assessed MP. SES was operationalized as poverty classification which was based on school SES. Significant small to moderate correlations (r = 0.19 to r = 0.36) were confirmed between early MP and later academic achievement, with impaired academic achievement and MP (p < 0.05) in low SES participants. MP, SES and academic performance in grade 1 were all significantly related. A direct association between academic achievement in grade 1 and grade 7 were confirmed. Academic achievement in grade 4 did however act as mediator and indirectly associated SES (- 0.413; - 0.417), MP (0.083; 0.093) and academic achievement in grade 1 (0.401; 0.398) with academic achievement in grade 7. Grade 4 academic achievement portrayed the largest standardized regression coefficients with grade 7 (0.693; 0.731). Low SES, and poor MP in first graders can contribute to compounded academic risk that are detrimental to academic achievement throughout the primary school years and should be addressed by timely interventions.