The key role of data mining in CRM

Enterprise Development CRM, the goal is two aspects, one is to help marketing staff manage their own sales process, the second is from customer data analysis of mining service development direction. The latter is the most important ...

Faced with brutal market competition, all enterprises are sparing no effort to win new customers. However, the existing old customers also contain huge business opportunities. The survey found that most companies have 20~50% customers every year, and this number is more in technology companies. On the one hand to strive for new customers, the other side but constantly lose old customers. To change this situation, retain old customers to win new customers, enterprises must fully tap the potential of existing customers. Through the customer's data mining to learn from the old customers, to explore new target customers, which is also a lot of successful enterprise CRM reasons. So a set of perfect CRM system, in the early stage of construction should seriously consider the need for data mining.

Demand and technology generate data mining

More common classification, CRM is divided into analytical, operational, collaborative, but no matter which one, the realization of the customer vivid understanding is the ultimate goal, so data mining in the core position of the CRM system.

Data mining is a "data generation" process for extracting useful information, it is to excavate the hidden, previously unknown, knowledge and rules which have potential value to the decision from a large amount of data, and can predict the result according to the existing information, and provide the basis for the enterprise's management decision and market planning.

The generation of data mining in terms of enterprise demand, after the launch of CRM, the operating characteristics of the first exposure, the company's day-to-day marketing business can be flow and automated management, and then the increasingly complex customer information, customer data accumulation, limited to marketing process management has been difficult to meet the further needs of enterprises, The entrepreneur expects the CRM to play the more important role, analyzes the massive complex customer data, the excavation customer value. CRM data should be adapted to a variety of analysis requirements:

No serious customer analysis, the face of the market can only blindly Siwang

Multi-dimensional analysis of customer characteristics: Mining customer personality needs, customer attribute description to include address, age, gender, income, occupation, education, and many other fields, can be multidimensional analysis of the combination, and quickly give a qualified customer list and number.

Customer behavior Analysis: Combine customer information to analyze the consumption behavior of a customer group. According to the different consumption behavior and its change, make personalized marketing strategy, and select "Gold Customer" from it.

Customer Focus Analysis: Customer contact and customer service analysis.

Customer loyalty Analysis: Customer durability, robustness and stability analysis.

Sales Analysis and sales forecast: including according to the product, the promotion effect, the sale channel, the sale method and so on analysis. At the same time, analyze different customer's different influence to enterprise benefit, analyze the influence of customer's behavior on enterprise's income, make enterprise and customer's relation and enterprise profit optimize.

Parameter adjustment: In order to improve the flexibility of the analysis results and expand the scope of application, enterprises need to adjust the relevant parameters. For example, what is the impact of price changes on income? What is the customer's point of consumption approaching and starting to become a "positive profit" customer? Enterprises need to collect all kinds of information for collation and analysis, the use of scientific methods to make a variety of decisions.

In addition, the development of information technology has made great contribution to the production of data mining. IDC's research Report, 2003 Data Warehouse will reach the market size of 20 billion U.S. dollars, data Warehouse is a decision-oriented, integrated by multiple data sources, with current and historical end data database system. It is a central storage system that helps enterprise employees answer business questions from their customers.

In CRM, Data Warehouse will be a large number of complex customer behavior data, set up a consolidated, structured data model, on the basis of data standardization, abstraction, standardized classification, analysis, for the enterprise management to provide timely decision-making information for the Enterprise Business Unit to provide effective feedback data. Now, NCR, IBM, Oracle and so on vendors are in the field of data warehousing, some predictive models and solutions have been established, data warehousing is not only a simple data storage, and become the customer data analysis, the cornerstone of mining customer potential.

Three stages of customer analysis

The customer analysis process includes the following three phases: Customer behavior analysis, key customer discovery, and effectiveness evaluation. First of all, the customer behavior data (feedback) and effectiveness evaluation results are focused on customer behavior analysis, through the excavation of key customers, to provide the basis for the formulation of market strategy, secondly, the analysis of customer behavior to the report form to market experts, market experts to use these analytical results to develop accurate and effective market strategy Finally, based on the market feedback provided by the customer, another performance evaluation is provided to improve the service and CRM itself.

1, Customer behavior analysis

Includes three steps for cross analysis between behavior grouping, customer understanding, and customer groups. Behavior grouping is the key, and the analysis result of behavior grouping makes the latter two steps easier.

Behavior grouping: According to the different customer behavior divides into the different group, each group has the obvious behavior characteristic. By grouping, we can better understand the customer and discover the behavior rule of the group customers. In the analysis process, the customer feedback received after a market activity is called "Reaction behavior Mode", and the "two-dollar customer response mode" adopted by the manual sales system is different, and the "classification reaction behavior pattern" adopted by CRM allows the definition of multiple reaction behaviors. The method of defining reaction behavior depends on the business area in which the enterprise is engaged. For example, the main business is clothing sales, a reaction behavior can be defined as "from the product catalog to buy women's clothing," can also be defined as "from the product catalog to buy men's clothing." The definition of these behavioral patterns can be very specific as needed (e.g., a red men's polo shirt).