The purpose of the present study was modeling villagers’ intention to adopt e-marketing and rural provincial clustering. The data was collected from 988 villages with ICT offices from 31 provinces of Iran. The results showed that the rate of agricultural income in villages with ICT offices was desirable but income of handicrafts and tourism was lower than the average. Also, rural access to city and ICT users were acceptable. In designing the research model, the “theory of planned behavior” (TPB) was developed. In this regard, a new factor called rural economic geography was added having five new variables such as rural access, rural tourism, agriculture, rural handicrafts, and population. Thus, the “Geographic Model of Planned Behavior” (GeoTPB) was presented. In the new model, the impact of geographic factor on behavioral factor was examined using structural equation modeling approach. In general, the rural economic geography had a direct impact (64%) on behavioral factors and in the next step, the behavioral factor (as full mediation) affected the intention. The Geographic Model of Planned Behavior predicted villagers' intention to e-marketing adoption up to 75%. In addition, the K-means clustering indicated that Iran rural regions can be placed in six clusters due to economical and behavioral factors. In most clusters, rural economic geography was effective on behavioral factor. The interesting finding was that Sistan and Baluchestan, Hormozgan, and Khuzestan provinces in cluster 1 were the leading regions in e-marketing adoptions. These regions were recorded as less-developed villages in the reports of the statistics center indicating that the development of technology was significantly effective on the digital literacy of the inhabitants of these regions. The findings emphasized the significance of considering behavioral factors in geographic studies to provide realistic solutions for rural development.