Abstract Study question Does fertility education improve semen quality in healthy men? Summary answer Healthy men who received online video-based education had significantly improved fertility knowledge and semen quality. What is known already Modifiable lifestyle and environmental factors can affect semen quality, but men generally have limited knowledge about fertility and are not aware of the impact of these factors on their reproductive health. Prior research indicates that fertility education increases fertility awareness, but whether this effect extends to modifications in lifestyle and thus to semen quality remains unexplored. Study design, size, duration Participants (N = 382) were randomised into two groups. The intervention group (IG) watched a five-minute online video series about healthy lifestyles and fertility risks (e.g., sedentary time, smoking, high temperature) every four months, a total of three times. At the same time points, the control group (CG) received a general health web article. Assessments of fertility knowledge, health behaviour, and semen quality were assessed before the intervention (T1) and one year later (T2) for both groups. Participants/materials, setting, methods Participants were male employees recruited during annual health check-ups at six workplaces in Japan. Those aged 20-39 years who planned to have children in the future participated. The primary short-term outcome was semen quality, measured by sperm concentration, sperm count, sperm motility, and DNA fragmentation index (DFI). Secondary outcomes were Cardiff Fertility Knowledge Scale (CFKS) scores and health behaviours (e.g., reduced sedentary time). All analyses were conducted on an intention-to-treat basis. Main results and the role of chance Of the 382 participants, 208 (55%) completed the T2 survey: 53% (99/188) in the IG and 56% (109/193) in the CG. T1 characteristics, fertility knowledge, health behaviours, and semen quality were well balanced between groups. Generalised estimating equation models showed a significant gain in fertility knowledge from T1 to T2 in both groups: +13.2 points on CFKS (95% confidence interval [CI]: 9.7 to 16.7) in the IG and +4.7 (1.3 to 8.1) in the CG, with a significant interaction between group and time (P<0.001). Total motile sperm count showed a significant interaction between group and time (P=0.046), with a significant improvement over time in the IG (+58.1 [15.2 to 101.1] million/ejaculate) but no change in the CG (-2.6 [-43.8 to 38.6]). DFI showed a marginally significant interaction (P=0.064) and an improvement over time in the IG (-0.72 [-1.54 to 0.10]) compared with the CG (+0.35 [-0.43 to 1.13]). The other semen indicators and health behaviours showed a favourable trend in the IG, but the interactions between group and time did not reach statistical significance. Limitations, reasons for caution Difficulties in recruiting participants and high attrition led to a small sample size, which may limit the statistical power to detect a significant link between knowledge gain and improved sperm quality via altered health behaviours. Volunteer bias towards men more interested in fertility may limit the generalisability of the findings. Wider implications of the findings Low-cost short video interventions could improve semen quality at the population level. As a future direction, the preventive effects of fertility education on infertility and the causal pathway should be explored in larger studies. Trial registration number UMIN000045389