Abstract: This study aimed to determine the effect of website features on relationship marketing among Hamsafar Travel Agency in Ahvaz City. The research method was descriptive, correlational of development-applied type. The statistical population consisted of all passengers of Hamsafar Travel Agency in Ahwaz who booked tickets online. Since, the size of statistical population of this research was unlimited, the sample size of 345 subjects was selected using randomized sampling. To analyze the data, to test reliability, Cronbach’s alpha coefficient method was used and value of this statistic for all variables was higher than 0.7, so questionnaire had desired reliability. To analyze the hypothesis, Pearson and Spearman correlation and regression models were applied using with SPSS Software and for structural equation modeling, AMOS Software was used. According to the results, order of dimensions from the the highest to the lowest mean value included affective commitment, calculative commitment, website features and relationship marketing. The results show that the website features had a significant positive impact on affective commitment and calculative commitment and that relationship marketing has a significant positive impact on affective commitment and calculative commitment has a positive and significant impact on relationship marketing. According to regression equations, website features accounted for 53% of the variations of affective commitment and 42% of calculative commitment while affective commitment accounted for 1% of variations of relationship marketing and calculative commitment accounted for 6% of variations of relationship marketing. In conclusion, it can be said that website features has a higher impact on affective commitment and calculative commitment has a higher impact on relationship marketing and calculative commitment and affective commitment have a weak mediating role between the website features and relationship marketing.