Biologically inspired models continue to drive the development of wide range of tools for understanding physical and social systems. Artificial neural networks and dendritic cell algorithms, to name a few, have been successfully utilized to aid in the development of tools and/or understanding of robotics [1], finance [2], artificial immunology [3], public opinion [4], differential equations [5, 6], and physics education [7]. Here, we demonstrate that dendritic growth model in a confined substrate can be used to capture the empirical features of multilevel marketing (MLM) schemes. Multilevel marketing is a form of direct selling where distribution and selling of products is administered through a direct contact between a buyer and an independent distributor [8]. In contrast with traditional media sources like television, radio, or newspaper, MLM makes use of a more direct word of mouth marketing to sell products and exploits the network connections of its members through recruitment processes. In recent years, the number of MLM companies is projected to continuously grow in the worldwide scene [9]. In the Philippines alone, the World Federation of Direct Selling Association (WFDSA) reported that the country has more than 100 MLM companies involving almost 4 million members as of 2014 with a combined direct selling retail sales of 1.2 billion US Dollars. Globally, WFDSA reported that the total sales of MLM companies for 2014 is about 180 billion US dollars, more than twice the sales of video gaming industry (∼$۷۶ billion) and about a dozen times of the music industry (∼$۱۵ billion). The MLM scheme naturally entices its stakeholders because it promises a personal financial stake for it consumers/inverstors, in terms of profit sharing and/or commission-based income [10]. However, MLM companies are closely watched by government agencies because of the similarity of their operating mechanism to illegal pyramid schemes [8]. The risk of MLM companies to degenerate into unethical and fraudulent pyramid schemes provides a compelling reason on why we need to better understand the mechanisms behind the growth of their networks. An accurate model of the connectivity of agents within the network is one of the most crucial step to advance this understanding. Previously, observed features such as parity, membership evolution, and Pareto distributed income are attributed to the fact that the networks underlying the simulations have scale free connectivity patterns and not random nor ordered [11].