Early stage startup would conduct ideation, problem solving, and market research. One of the stages in market research is market segmentation. Normally, early stage startups do not have enough resources, thus many processes are done manually. This encourages inacurracies and lessen the objectiveness in evaluating market situation which is important for the growth of early stage startups. This research focuses on developing a machine learning based application for market segmentation. The framework used here is CRISP-DM, which is a framework used in data mining. This framework has six phases, which consists of business understanding, data understanding, data preparation, modelling, deployment, and evaluation, to identify the input and output of a process. The data model used in this application is K-Means, which is a common algorithm used for clustering or dividing a set of data into groups according to their attributes. The application developed is able to output results in the form of visualization and the segmentation in excel format. With this, the early stage startup is able to process their data to identify segments in the market. For future research, this application could be improved in the efficiency and accuracy of the model. The application could be improved in the aspects of UI/UX design, the algorithm used, the number of clusters, and the analysis of the dataset, as well as adding a predictive analysis feature