5.2 Customer Intelligence and Data Mining

The development of direct marketing along with the application of e-business and BI technologies resulted in the emergence of a subdiscipline of marketing based on client analysis, which is called customer intelligence (CI). One may interpret this as the analytical processing of data about a client for marketing purposes. This approach entails the collection, analysis, and use of data about a client. The client’s behavioral history is investigated in order to determine his or her profile, preferences, and needs. This serves as the basis for the creation of a marketing message customized for a specific client. The main areas of support for marketing client analysis by BI techniques within CI are as follows:

1. The identification of potential clients (target group)
2. The selection of communication channels
3. The choice of a relevant message and time to convey it

Each of these areas may be supported by the application of data-mining methods. These techniques, which allow the possibility of classification, enjoy considerable popularity. It is possible to build precise models of either communication channel selection or the choice of the message itself. It is worth mentioning that there are methods that enable one to compute the correspondence between a real client profile (a de facto recognized profile) and the pattern profile of the client who is a purchaser of a given product. It permits one to carry out an efficient selection of clients and recipients of a given advertising campaign, and it delivers adequate results. The campaign is targeted at clients who correspond to the pattern profile (in the context of similarity measure).⁶ The advertising campaign is then reviewed by a response model, which enables you to prove the hypothesis that there is a higher response from clients who are selected according to a correspondence with the pattern profile compared with clients selected at random.⁷

To illustrate the issues discussed, let us consider the following examples of their application:

1. Selling cosmetics in perfume chain stores
   • Client identification. It was determined that the client is keen on beauty care (information obtained from analysis of posts on the Internet forum), and her profile meets the requirements of the following pattern profile: sex = woman, and age = 25+.
   • Channel of communication. Short message service (SMS) or multimedia messaging service (MMS) can be used, depending on the model of the client’s phone.
   • Time of message. The message can be transmitted when the client enters a shopping mall.
   • Message. The information will be about a special offer of a new line of cosmetics in a store located in a shopping mall.
2. Selling of financial products by a retail bank
   • Client identification. The pattern profile consists of clients with a high income, in managerial positions, age = 40+. Purchasers of cars make α model β were observed to have a corresponding profile. Thanks to an agreement with a car dealer of make α, access to the e-mail addresses of customers of model β was obtained.
   • Channel of communication. E-mail.
   • Time of message. Simultaneously with an invitation to join an enthusiast club for owners of car make α.
   • Message. Personalized offer of a long-term deposit account.