Integrate Data Mining, Campaign ManagementDatabase marketers stand to gain competitive advantage by harnessing two merging technologies. Companies that successfully integrate data mining technology and campaign management software will:
* Improve the accuracy and results of marketing campaigns.
* Shrink direct marketing cycle times.
* Increase knowledge of customers and prospects, which, in turn, will enhance companies' abilities to refine future marketing campaigns and fuel customer relationship management strategies.
What is data mining? Today, companies use data mining software to analyze large volumes of information to predict the behavior of customers and prospects. For example, a statistical analyst might use the software to build a model that predicts a customer's likelihood to "defect" from the company, purchase products across product lines or respond to an offer for free merchandise. This statistical model runs against a customer database, assigning a score that indicates each individual's or household's probability of exhibiting a particular behavior.
What is campaign management? Campaign management software automates the planning, execution and assessment of highly targeted marketing promotions. It directs timely, pertinent communications to customers across multiple "touch points," or channels. Done correctly, campaign management software will manage customer relationships across their life cycles through triggers that respond to timed events as well as customer action or inaction. For example, a customer who fails to use his bank home credit line for a period of three months might trigger a direct mailer suggesting uses for the line of credit.
Why integrate mining and campaign? Companies that apply data mining results to their marketing campaigns can further refine already-defined market segments. For example, a cataloger plans to send an expensive new catalog to high-income households interested in glass art. A statistical model determines households most likely to order from the catalog. By selecting only the top model scores of households within the targeted segment, marketers can eliminate customers and prospects unlikely to respond, sparing production and mailing expenses, and increase response rates and return on investment.
Unfortunately, the use of modeling scores in direct marketing campaigns remains a manual, error-prone and time-consuming process. As companies move toward targeted marketing, they must fire off more campaigns at short and repeated time intervals. For example, a company might execute a campaign to new customers daily and a campaign to customers approaching their renewal dates weekly. Marketing and technical professionals chartered with applying model scores to these campaigns may be hard pressed to keep pace with this pace.
The integration of data mining and campaign management eliminates the need to score an entire database. Instead, statistical models score only customers or prospects within a defined market segment. A tightly linked, automated process also reduces manual intervention and human errors, increasing the accuracy of campaigns and accelerating the direct marketing cycle. Marketing acceleration lets companies reach customers and prospects ahead of competitors and provides increased opportunities to learn from campaign results, each cycle turn represents another opportunity to refine marketing offers and efforts.
From the perspective of a database marketer, here's how the tight linking of these two technologies might work: Near the time a promotion is scheduled to run, a campaign management application calls an already-built statistical model from a library of available models to score a particular market segment. The application automatically determines which records within the database to score and when to score them, ensuring that the latest information is used. By using statistical models within the context of campaign management, a marketer gains a simple process to schedule model scoring, which can keep pace with campaigns that run frequently.
A business-driven solution. This integration is not technologically in search of a solution. It is a business opportunity being driven by companies such as Fleet Bank, who recognize the advantages that can be gained from such a union. Fleet will integrate SAS Institute's new Enterprise Miner product with its campaign management system.
Today, data mining software can help find the "high-profit" gems buried in mountains of information within a database that contains hundreds of thousands or millions of customers.
"However, merely identifying your best prospects is not enough to improve customer value," said Randall Grossman, Fleet's vice president of database marketing. "You must somehow fit your data mining results into the execution of marketing campaigns that enhance the profitability of customer relationships."
Data mining and campaign management technologies evolved along separate paths -- until now. Organizations stand to gain a competitive edge by understanding and using this new union.
Andy Frawley is president/CEO of Exchange Applications Inc., Boston.