Data mining for the profitable veins of gold

It's no news that maintaining strong customer relationships is a key driver in getting more value out of a growing business.

But as a business acquires more customers, many firms find it much harder to answer basic customer-centric questions, such as: who are our customers? Do they buy a little or a lot? Is that once a week or once a year? How do they use our products or services? What is their preferred method of communication?

Not being able to answer these questions means businesses are not getting the most value out of their interactions with their customers.

Customer Relationship Management (CRM) tools are one way that many companies address this issue, which is why it has become one of the hottest growth segments of the software industry.

But this can be too expensive for smaller firms. Another equally effective and often cheaper way to manage customer relationships is data mining existing commercial databases and information.

Every day, a growing business will process and store vast amounts of highly detailed information about customers, markets, products and processes - whether it is through bookkeeping programmes, which contain direct sales data; server information about online sales; or in the stack of business cards which you're going to file "when I have time".

Data-mining this information gives businesses with lots of customers, or lots of customer information, the ability to make knowledge-driven strategic business decisions to help predict future trends and behaviours and create new opportunities.

While data-mining can be done on pretty much any type of customer information, the first practical step that a business can take to get a better understanding of its customers to is to collect good data - that is to record, measure and learn.

A good way to start the recording and measuring process is to apply a direct marketing discipline called RFM - or recency/frequency/monetary Value.

"Recency" is the time since you last saw the customer; "frequency" applies to how often you see them; and "monetary value" applies to the amount they generate for your business - revenue or profit.

The way this is applied will vary between industries but the data that is collected serves the same purpose - it helps drive value in the business.

For example, in the airline industry, a customer may fly only a couple of times a year, but always fly business class. Understanding the behaviour and preferences of a customer segment helps the airline direct-market to that customer more effectively and put the products and services in place that will better cater to their needs. The end result is, hopefully, a more loyal and higher value customer relationship.

For a smaller firm, putting out a flyer about a new product or service also provides a good chance to collect information about customers. Recording and measuring the uptake of the product or service, who the customers were, and what else they bought will help to drive more effective future marketing initiatives.

And the information collected for data-mining is not only useful from a marketing perspective. Data about customer behaviour and preferences should also feed into strategic business decisions about new investments in such productivity driving areas as training and new products and services. All extract more value from existing business processes.