The American mobile telecommunications industry has matured from its early days of explosive growth to a saturated and fiercely competitive market where carriers are battling to simply retain existing customers. The industry now has one of the highest customer churn rates in business; as an example, in August 2012, T-Mobile US’ churn rate was 2.1%, which may seem miniscule until you factor in its millions of subscribers. In fact, voluntary customer churn has become so pervasive that many companies are redirecting vast sums of their marketing dollars from developing customer acquisition programs to devising churn management strategies. Gartner recently reported that marketing budgets for customer retention will increase by more than 50% by 2015.
Most often, service providers are reactionary in their efforts to address voluntary churn. But what if unstructured data could be used to develop a model that could identify the most lucrative customers and help develop a proactive program for ensuring their loyalty?
At the end of the day, the goal for any voluntary churn reduction program is revenue stabilization and augmentation. Given that, looking at a subscriber’s profitability is arguably a better model for developing a targeted marketing program than utilizing overall subscription numbers as a benchmark. Not all subscribers are equal in value; in fact, based on cost versus profitability, there are some customers that it might be wiser for a service provider to proactively lose from a revenue standpoint.
The Alacer Group tested this theory in a recent project for a U.S.-based tier-one service provider. By mining unstructured historical data to understand and track the variables that caused its customers to leave, we developed a model that could identify the most profitable customers vulnerable to future churning. Data was extracted from activity areas such as customer billing, network access and call detail records. We could then predict the level and likelihood of voluntary customer churn – giving the carrier a proactive roadmap for countering it to help maintain profitability.
The objective was to sift through enough data to develop a plan to proactively reach out to subscribers before they terminated their contracts. In taking a profit-centric approach to the task, we developed a profit-churn score for each customer by examining more than 70 different pieces of information on more than 70,000 subscribers. Through a binary logistic regression, we could predict with a 60% accuracy rate whether or not any given subscriber would churn. Each customer was then assigned to four different quadrants within a matrix that described the nature of the customer and helped determine actions to be taken for desired outcomes.
And this is where it got interesting: we learned that 35% of the carrier’s subscribers were more profitable than the average customer, and that they generated a whopping 61% of the carrier’s cumulative expected profit per month.
To identify these highly profitable subscribers, each received a combined score to assign them to a specific quadrant. For example, a customer with a profitability score of 7 and a churn score of 5 received a combined score of 75. Each two-digit score could then be read like a set of X,Y coordinates. The boundaries of the quadrants were created from carrier data set at break points driven by business needs. The churn axis breaks between 2 and 3 where there is a 50% or more probability of churning. The profitability axis breaks between 3 and 4 where the mean profitability was calculated.
By putting the churn and profitability models together, we end up with a useful framework for informed marketing actions. The profitability of each customer can be used to prioritize those who should receive immediate attention, and identify the high cost/low return customers where churn might actually be desirable. The carrier can now develop retention plans focused on its most highly profitable customers who are likely to churn; unprofitable customers are either allowed to leave or targeted for movement to a more profitable status.
The model identified several churn predictors and potential strategies that could be used to retain the carrier’s most valued customers. Examples include:
–Customer service calls. In this instance, the customer has contacted the customer service team. Through an integrated CRM program, the team can indicate that this customer is a retention priority. The profitable customers who are likely to churn would be at the top of the work queues for callbacks.
–Higher unique subscribers. This would indicate that the customer does a lot of business with the carrier and will regularly shop for volume discounts. The carrier could proactively discount additional lines and offer to consolidate any lines the customer has across multiple carriers.
–Children in household. The carrier could offer discounted family plans to encourage others in the household to join the existing plan instead of choosing a competing offer.
–High credit rating. These are customers who can easily take business elsewhere since they are not locked into a plan due to poor credit. For these customers, the carrier can proactively offer early discounts to extend a 24-month commitment.
A quantitative approach to combining churn and profitability models helped this carrier develop voluntary churn counter measures that effectively moved it from a reactive stance to a proactive one. It’s just one example of how big data can provide marketers with a more useful framework for developing proactive action plans that positively affect the bottom line.
Reality Check: Using profitability to determine churn strategies by Ed Sarausad
RCRWireless, June 11, 2013
