In the current complex and dynamic economic environment, it’s crucial for enterprises not only to respond to external market changes but also to fully utilize and deeply analyze their own business data. This approach helps in promptly identifying opportunities and risks, thereby achieving long-term sustainable growth. Cash flow indicates the inflow and outflow of funds within a certain operating cycle of an enterprise, reflecting operational efficiency and development trend, and has attracted increasing attention from academia in recent years. This study constructs a multivariate Hawkes process model based on tens of thousands of cash flow transaction records from 67 chain pharmacy enterprises. From the perspective of cash flow transactions, by combining model interpretation with real case analysis, the study explores internal influencing factors of cash flow data and conducts in-depth analysis for different business categories. The study finds out: Parameter analysis of the amount-binned multivariate Hawkes model reveals that events with larger inflow amounts generally have greater spontaneity, small amount events have stronger self-excitation effects, and mutual excitation among small-amount events is also stronger; Parameter analysis of the business-type multivariate Hawkes model can identify cash inflow businesses with high spontaneity such as loan income, investment income, inter-company borrowing and internal transfers, as well as inflow businesses with strong excitation effects such as cash deposits and sales inflow. This can help decision makers better identify the main operating models of related companies, adjust strategic planning, and improve cash flow management.