Enterprise information is represented by artifacts shaped by applications, projects, technologies, organizational boundaries, and local requirements. Over time these structures accumulate, creating representational complexity that the enterprise must maintain and that information consumers and AI systems must interpret. This paper introduces Enterprise Representation Simplification (ERS), which aims to remove unnecessary representational complexity within a defined scope while preserving required information. To compare complexity across representation states, the Enterprise Representation Complexity (ERC) model is defined using four dimensions: Representation Objects, Interactions, Behaviors, and Supporting Sources. Objects, Interactions, and Behaviors are inter‑dependent, while Supporting Sources describe the exposure of the representation. ERC can be defined at both the representation level and the task level, enabling comparison of alternatives and distinguishing architectural simplification from retrieval optimization.
Two direct consequences of ERC are developed. First, representational structures create lifecycle obligations such as maintenance, governance, dependency management, change, enhancement, and operation. An economic model partitions cost into recurring global representation cost, recurring task‑level cost, and one‑time transformation cost, allowing evaluation over a specified time horizon. Second, reducing task‑level ERC narrows the representational extent an AI system must identify, relate, and interpret. Evidence from Text‑to‑SQL research shows that simplifying schema and reasoning complexity can improve reasoning accuracy. It is important to note that ERC is not a universal metric of complexity, performance, or cost; rather, it provides measurable architectural variables for comparing representation alternatives, transformation effects, economic outcomes, and AI reasoning performance.
Review: ERC offers a systematic framework for enterprises to assess and prune representational redundancy while preserving business integrity. Coupled with the economic model, it enables decision makers to quantify simplification benefits, which is especially valuable in enterprise AI scenarios driven by large language models.