Data Mining and Marketing

HOW TO INCREASE REVENUES AND PROFITS FROM MARKETING?

Nowadays marketing and computer technologies provide unlimited possibilities for collecting and storing data from interviews, surveys and other sources. This could be precious information, that can help with increasing campaigns response rate, improving customers relationships management(CRM), reducing marketing campaigns costs, etc.

For example, to be successful, companies must be proactive and predict what a customer needs. Customer profiling provides the basis for starting what marketers call a "dialogue" with customers. Arranging customers classes allows to increase response rate for direct marketing campaigns, targeting a direct marketing campaign to similar individuals.

But is it possible to manually deal with vast amounts of data, collected in databases and to quickly respond to customers wants and needs? How to receive the most of information in minimal time?
When you start asking such questions, it is time to get acquainted with data mining software more closely.

The key is to ask the right question. Lets take a direct marketing campaign as an example.

These are the questions, that can be asked:

What kind of data can be helpful for clients profiling?

To profile customers, you can use your current customers database or results of surveys / interviews. This can be information about their purchases, hobbies, everyday needs, and of course, more personal information about their age, sex, marital status, personal income, etc.

How should be the customers data stored for receiving best analysis results?

The best way to store customers data are structured databases. This will allow to easily use the most important profile parameters during the data mining process. For example, you might be interested in targeting your company for a certain age and sex, in this case the database should contain such information. Besides, the database should contain the key information, for example: is client interested in your product (yes / no / not sure / don't know about the product, etc), clients attitude to the kind of advertising you are going to use (negative / positive / neutral).

After presuming the customers classes, it is important to create a database which will contain information about all representatives of customers groups. Only in such case you will receive a summary of portraits for all groups.
 

How can clients be classified (profiled)?

Classes should be created depending on the object of the customers database analysis.
If you tend to unearth the portrait of a customer, who is interested in a certain product, you should set parameter(database table field), containing information about it , as the class field. If we return to the example above, we will receive 4 classes:

• yes
• no
• not sure
• don't know about the product

Combining obtained data with an analysis of clients attitude to advertisement types will allow to create a targeted marketing campaign, based on results of data mining.

How to apply the results of data mining?

Now, after data mining the customers database, the new information can be used not only for targeted marketing campaigns, but for fast interaction with each new client of your company. After receiving some information about a new client, you will be able to classify the customer, predict his expectations and needs for mutual understanding between customer and company. As a result you increase the probability of retaining customers in the face of competition.

Examples of data mining application in marketing

Banks use new database marketing techniques to identify the best customers. This allows to target customers for loan campaigns, to forecast customer retention and to track direct marketing campaigns.
Many telephone companies use data mining for analysing customers calling patterns. As a result, they are able to recommend the best plan for each new client from the very beginning of interaction.

Database marketing analysis procedures and techniques are a must for keeping up with the vast amounts of data now available. The data mining technique is not hard to understand and use, especially keeping in mind the precious data on customers, customer buying and behaviour patterns, and other valuable information that can have a fundamental impact on companies revenues and profits.

July 2007,

H. Smirnov-M (c)