IBM SPSS Statistics: How to use control packaging test to optimize marketing results

IBM SPSS Statistics http://www.aliyun.com/zixun/aggregation/10341.html "> Statistical analysis Forecasting Software is a survey in the statistical industry, market research, medical statistics, Government and enterprise data Analysis application of the long-established statistical analysis tool, is the world's earliest statistical analysis software, in the global has a large number of enterprises and individual users.

Although the statistical analysis of IBM SPSS Statistics is powerful, it needs to rely on the rich background knowledge of statistical analysis to operate, so it is inconvenient to use by commercial users. To this end, after the 18 version, the IBM SPSS Statistics new Customer Direct Sales module, the module's operating interface is simple and clear, the results of the report analysis clearly understandable, can be widely used in telecommunications, retail, banking, insurance, securities, media, market research and other industry fields, It is a tool designed for marketing personnel to improve the efficiency of direct selling and improve the effectiveness of direct marketing activities.

Open the IBM SPSS Statistics, choose "Direct Selling"-> "Select Method" in the menu, enter "Direct Selling" module Select Method interface, as shown in Figure 1.

Figure 1. Direct Marketing Module Selection method interface

In the "Direct Sales" module, there are three parts:

understands my contacts, which are used to analyze customer information and classify customers according to different characteristics. Improve my marketing activities to predict customer responsiveness to marketing activities. Scoring my data, using the "choose the most likely to buy contacts" and a variety of programs in other modules to build a predictive model to score new customer data based on the forecast model.

This article will introduce the second block of the direct marketing module, "Improving My Marketing Campaign":

comparison activity effect (control packaging test)

With this article, you can learn how to use this feature in the Customer Direct analysis module to improve marketing activities to help you provide a more reliable basis for developing Smart marketing strategies for different customers.

How to use control packaging test to optimize marketing effect

Control packaging inspection is used to compare marketing activities to check whether there is a significant difference between the different packaging or commodity. Help marketers determine which package (product) will have better results.

Application Background and Data description

A company has designed a new packaging for a product, and the marketing department wants to test whether the new packaging will get a higher customer response than the existing packaging by sending a message. The test message is sent to two customer groups: control group and test group. The customer of the control group will get the existing packaged product, and the test group will get the new packaged product. Then, the results of the two groups were compared to check whether there were significant differences. Choose the most popular new packaging based on the results.

The customer's data information is stored in the SAV format of the IBM SPSS Statistics storage file, as shown in Figure 2, it contains the following fields: Customer ID, customer response to the test product sent (yes, express willingness to purchase this product, no, no intention to purchase), control package (0, The control group, 1, indicates the test group. Postal code, age, income, educational level, residence, sex, marital, area, etc.

Figure 2. Control Packaging Inspection Customer information file variable view

Data preparation

Controls the wrapper field, which must be a category variable to represent different activities, such as packing A, packing B. The type of response variable that the customer responds to should be a category variable, and there is no order, which is classified as a nominal nominal variable. A value for the variable represents a positive response from the customer (for example, the customer is willing to buy the product), and other non-missing values indicate a customer's negative response (for example, the customer is not willing to buy the product). If the response field is a continuous value (Scale variable), such as the customer's purchase quantity or purchase amount, you need to create a new variable to specify a unique positive response value for all non 0 response values.

The definitions and settings for variable metrics can be found in the reference section of the IBM SPSS Statistics Introduction and the basic concept introduction in the "Data Prediction statistical analysis product IBM SPSS Statistics Example Application explanation".