This study was designed to determine the effects of twinning on the subsequent reproductive performance and productive lifespan of high milk-producing dairy cows. The data analyzed were taken from complete reproductive records for 4861 Holstein Friesian cows comprising a commercial herd, including data for 12 587 calving events from April 1, 2001 to March 31, 2012. The twinning rate was 5.6%, and 9.6% of the cows delivered twins at least once during their life. Conception rates before Days 90, 120, and 300 postpartum were 20.8%, 36.9%, and 74.5%, and 34.2%, 51.8%, and 85.0% for cows delivering twins and singletons, respectively. Cox regression analysis revealed that twinning reduced the chances of conception before Day 90 by a factor of 0.76. Also, the median calving to conception interval was significantly longer for cows calving twins (134 ± 4.5 days) than for cows delivering singletons (108 ± 0.8 days; P < 0.001). Moreover, in cows in which conception was successful, the risk of abortion was higher for cows that calved twins than for those calving singletons (13.7% vs. 10.3%, respectively; P = 0.01). Culling rates before Days 90, 120, and 300 postpartum were 15.6%, 16.1%, and 28.6%, and 7.6%, 8.7%, and 15.9% for cows calving twins and singletons, respectively. Cox regression analysis of the factors affecting the culling rate before Day 300 postpartum revealed a 1.41 times greater hazard ratio of culling for cows calving twins than cows calving singletons. Further, mean productive lifespan was almost 300 days shorter for primiparous twinners (N = 48, 602 ± 493 days) than for nontwinners (N = 2592; 899 ± 581 days; P < 0.01), and 200 days shorter for secundiparous twinners (N = 126, 914 ± 429 days) than for nontwinners undergoing at least two lactations (N = 1936, 1101 ± 522 days; P < 0.01). Kaplan-Meier survival curves for productive lifespan differed between primiparous twinners and non-twinners (P < 0.001), and between secundiparous twinners and nontwinners having at least two lactations (P = 0.017). Differences in culling patterns for twinners and nontwinners were not restricted to the subsequent lactation but continued as long as 800 days after first calving, strongly suggesting long-term negative effects of twinning. As an economic implication of our findings, we suggest that twin embryo reduction at the moment of pregnancy diagnosis could be a profitable strategy to cut twinning rates and abolish their detrimental effects on subsequent reproductive performance and productive lifespan.