Data-Empowered Multi-level Weighted Rule Evaluation for Customer Classification in the Tobacco Industry—A Foundation for Market Status Identification and Predictive Regulation

Abstract

Changes in consumer spending patterns have a significant impact on the market, and grasping consumption trends is crucial to the development of enterprises. Taking Liangshan tobacco marketing as the research object, this paper analyzes retail business data to classify retailers into different types, capturing changes in cigarette consumption to support tobacco marketing decisions. By establishing a retailer quality scoring model, retailers are evaluated from four dimensions: basic compliance, data accuracy, sales behavior normativity, and initiative, ultimately forming a comprehensive evaluation of retailers. Based on the evaluation results, retailers are classified into four categories: Benchmark, Stable, Follower, and Lagging. This classification provides an effective evaluation standard for subsequent marketing decisions, enables efficient resource allocation, and encourages retailers to actively respond to market changes, thereby enhancing customer satisfaction.