Children are unique and potential market. Unique, because children usually do not spend their own money but they have huge purchasing power. Potential, because from the number of children and spending, they are big and promising. Children market consists of three markets, they are: First, primary markets which targeting the children itself. Second, influence market which emphasized the influence that children assert on family purchases. Third, future market which refers to the potential market from the children in the future. Although children are huge markets and consists of three markets, however research which see children as a combination of three markets and using children as object in research are still limited. Researches are still concerning in children as primary markets and using parents as source of information. This research aims to understand the three markets of children together with children as source of information. The writer wants to identify the antecedents of intention to consume for children and its impact on influence power, impulsive buying, and autobiographical memory. Furthermore, it is needed to identify the antecedents of autobiographical memory for children. Using Structural Equation Modelling, the researcher wants to know the relationship for each variable simultaneously. Using cluster sampling, this research will be conducted in Jakarta and Semarang with total 468 respondents which consists of 204 state elementary students which are taken from ten state elementary schools in Semarang and 264 private elementary students which are taken from ten private elementary schools from five regencies in Jakarta. Private schools in Jakarta represent low involvement consumers whereas state schools in Semarang represent high involvement consumers. The results show that promotion influences intention to consume, whereas product characteristics, reference groups and retail environments do not influence intention to consume significantly. Furthermore, results also show that intention to consume influences autobiographical memory whereas influence power, and impulsive buying do not influence autobiographical memory significantly. The results also show that impulsive buying influences to influence power, whereas intention to consume does not influence significantly to influence power. From the result it could be noted that service providers have to conduct advertising' activities which directly target children, fulfill its promises, and communicate the advantages of products and retail environments which have been developed. They have also to conduct event marketing which is innovative, oriented to children, continuously, and flexible. The next implication is about educating staffs to be proactive in order to encourage children to buy by their own
