A multidisciplinary Australian University introduced a block model of blended, sequential 4-week blocks to first-year students. This natural experiment compares the inaugural block and two prior cohorts on satisfaction and performance (n = 15,989 satisfaction and n = 86,545 assessment observations). Mixed effect cross-classified models with comprehensive controls, moderation and sensitivity testing show substantial increases in performance, especially for equity groups important in an expanding sector: low-socio-economic and prior academic achievement, non-English-speaking background. Effects on satisfaction were small with increases in teaching satisfaction but decrement in course satisfaction, especially perceived reasonable workload. Discipline consistently moderated effects and units that redesigned assessments offset some decline in course satisfaction. New models of higher education may support improved outcomes.