This randomized, controlled clinical trial examined the impact of a web-based weight loss intervention on diet quality. Furthermore, it was investigated whether corresponding changes in diet quality were associated with changes in measures of cardiovascular risk profile. Individuals with a body mass index (BMI) of 27.5 to 34.9 kg/m2 and an age of 18 to 65 y were assigned to either an interactive and fully automated web-based weight loss program focusing on dietary energy density (intervention) or a non-interactive web-based weight loss program (control). Examinations were performed at baseline (t0), after the 12-week web-based intervention (t1), and after an additional 6 (t2) and 12 months (t3). Based on a dietary record, the Healthy Eating Index-NVS (HEI-NVS) was calculated and analyzed using a robust linear mixed model. In addition, bootstrapped correlations were performed independently of study group to examine associations between change in HEI-NVS and change in dietary, anthropometric, and cardiometabolic variables. A total of n = 153 participants with a mean BMI of 30.71 kg/m2 (SD 2.13) and an average age of 48.92 y (SD 11.17) were included in the study. HEI-NVS improved significantly in the intervention group from baseline (t0) to t2 (p = 0.003) and to t3 (p = 0.037), whereby the course was significantly different up to t2 (p = 0.013) and not significantly different up to t3 (p = 0.054) compared to the control group. Independent of study group, there was a significant negative association between change in HEI-NVS and dietary energy density. A higher total score in HEI-NVS did not correlate with improvements in cardiovascular risk profile. The interactive and fully automated web-based weight loss program improved diet quality. Independent of study group, changes in HEI-NVS correlated with changes in energy density, but there was no association between improvements in HEI-NVS and improvements in cardiovascular risk profile.