Enterprise Marketing Resource Optimization And Analysis Based On Big Data

The distribution of marketing resources and business development strategy is closely related to the future of enterprise.However,currently individual subjective judgments play an important role in marketing resource allocation.It's necessary to increase the market factors in order to provide scientific decision support.This paper proposes an architecture of data collection and analysis based on mass market data.According the collected information of national market,such as product sales,structure,region.It completes the storage and processing of market data.It extracts the decision variables of marketing resource allocation and combines with enterprise customer classification.It uses the brand life cycle management as the core and establish the marketing resources allocation model.It is based on efficiency,market trends and other evaluation basis,seeking to the optimal scheme to achieve scientific distribution of marketing resources and support the refinement management of enterprise and dynamic decision-making.The specific contents of this paper are as follows:1)Perception of market data and the collection of the enterprise massive data through market visits,timing input,terminal acquisition and others ways.Data that can be collected through internal reporting includes planning data,order data,customer data,and so on.Data that can be collected through the external data includes retail order data,logistics data,inventory data,and other business data.Based on the massive data acquisition,the enterprise data structure is constructed,including data analysis,processing and display.2)In the marketing resource allocation process,there are many factors,including market factors,geographical factors,corporate positioning factors.This paper Uses the random forest to extract the decision variables of the marketing resource allocation of marketing enterprises.Combined with the large-scale and time-sensitive characteristics of the enterprise data,this paper proposes a customer classification model based on hierarchical clustering and k-means clustering to segment the regional market.3)According to the needs of different stages of the life cycle,regional markets have differentiation on the demand of resources.Combined with enterprise marketing architecture,this paper puts forward the model of marketing resource allocation.Through the investment of resources,the brand can get more customers and quantity in the corresponding life cycle,increase product sales,prolong product life,maximize enterprise benefit and enhance the efficiency and competitiveness of enterprises.4)Verifying the scientific rationality of resource allocation,correcting and validating the marketing resource allocation model through sales forecast.In the process of sales forecasting,three kinds of sales forecasting models are established,such as linear regression,AR model and BP neural network.Through analyzing and comparing the prediction results,the prediction model with high accuracy is selected to modify the coefficient of marketing resource allocation model.5)Building the enterprise's marketing resources centralized platform and putting the marketing resource allocation model into actual use.By comparing with the efficiency of traditional marketing resource allocation model,it is proved that the marketing resource allocation model based on massive data can help enterprises optimize the marketing resource allocation process,improve the efficiency of marketing resources and improve the sales profit of enterprises.