This is often because organisations underestimate the complexity of the issue and attack individual problems or 'symptoms' instead of assessing the entire organisation and applying a comprehensive, integrated churn management strategy, suggests Knowledge Factory.
Customer churn - the propensity of customers to cease doing business with a company in a given time period - has become a significant problem that spares no industry. Although particularly rife in industries affected by deregulation or fierce competition, churn also affects organisations with aging customer bases or niche market products.
"Churn rates are soaring in a wide variety of industries, from telecommunications and financial services to publishing," explains Suben Moodley, senior analyst at Knowledge Factory.
Citing factors like the legislation affecting data privacy in the insurance industry and the mobile number portability (MNP) regulations, due to be introduced later this year, Moodley emphasises that churn is growing into an ever more pressing concern. "Wherever you look, the legislative and regulatory signs are that it could prove a very expensive concern to ignore," he confirms.
Predicting customer churn and designing cost-effective strategies to reduce it are not easy tasks. Most organisations know what their churn rates are, but are less clear about the specific underlying factors affecting the rates. As a result, they often adopt a 'shotgun' approach to the problem and run large promotional campaigns against the customer base that are rarely effective.
"Even if a company has a reasonable idea of what is affecting their churn rates, these problems rarely receive enough strategic attention," maintains Moodley. "Our experience has taught us that typical churn reduction initiatives consist of several disparate, functionally-focussed activities which attack the symptoms of churn rather than its root causes."
He uses a typical customer service initiative to illustrate the point: "An overloaded call centre may appear to be frustrating customers and causing churn, but reducing wait times by recruiting more staff won't resolve the issue if call volumes are high because sales people aren't educating clients properly about the service or product when they sign up."
"The first step towards managing customer churn effectively is to devise strategies based on sound empirical evidence and strong analytics," explains Moodley. "Companies need to use predictive customer churn modelling methods to extract insights from their existing business intelligence activities."
By designing and combining powerful data mining, data enrichment and predictive modelling techniques, companies can identify and respond to the root causes of their churn rates.
"Not only are organisations able to target their efforts more accurately and cost-effectively, but they can respond to churn much quicker," asserts Moodley. "Such an approach allows for trigger-based decision making in reaction to customer activity."
Nonetheless, managing customer churn is inherently cross-functional and Moodley is quick to point out that data mining alone will not help an organisation retain customers anymore than shotgun promotions will.
"Churn reduction, particularly with an emphasis on long term profitability, requires the integration of technology and technical analysis with sales and marketing and customer facing initiatives," he explains. "Many initiatives fail to deliver because the high level strategic support is not there; making it very difficult to set overall priorities and coordinate the efforts of competing issues and departments."
"An integrated churn management strategy needs comprehensive market and marketing datasets and assistance from a company that has experience in both the building of spatial and statistical models and data mining", says Moodley.
"But churn is a managerial as well as statistical challenge," emphasises Moodley. "The sales and marketing, finance and operational functions must collaborate to deliver on the strategy and bring the revolving door of customer churn to a standstill," he concludes.
