Although significant advances have been made in the last two decades regarding the use of dietary fibers to manipulate colonic microbiota composition towards a healthy state, relatively little progress has been achieved to do this in a predicted way. Part of the problem is that there is a lack of knowledge in understanding dietary fiber requirements of individual bacteria or bacterial groups residing in the colon. Dietary fibers are composed of an extremely diverse range of chemical and physical structures. Dietary fiber utilization abilities of microorganisms are controlled by gene clusters that consist of degradation enzymes, binding proteins and transporters. Dietary fiber utilization systems of bacteria must be integrated into a broader context of the highly competitive environment of the colon. Accordingly, fiber substrates must be chosen that both align with bacterial requirements and satisfy their ability to compete for them. In an attempt to create a better conceptualization of how fiber structures might favor individual bacteria or bacterial groups, we put forward a view of dietary fiber as a vast collection of “discrete structures” that are defined as unique chemical and/or physical structures which align with encoded gene clusters in bacterial genomes. In this presentation, we will present our perspective of how fibers could be selected to drive shifts in the colonic microbiota for better health.