Friends' influence helps telecom firms retain customers

Retaining customers is a central concern in many industries, including IT markets, where churn rates–the rate at which subscribers to a service discontinue their subscriptions–are high. For example, cell phone companies report churn rates of as much as 2% per month, which affects companies' value and profitability. A new study tested a strategy to help a telecommunications firm manage churn. Contacting not only customers but also their friends helped reduce the customers' propensity to discontinue their ties to the company.

The study, by researchers at Carnegie Mellon University, Católica Lisbon School of Business and Economics, and Erasmus University, is published in Marketing Science.

<p>&quot;Our findings suggest that firms can reduce churn by contacting customers&#039; friends even if the friends aren&#039;t likely to churn,&quot; explains Pedro Ferreira, associate professor of information systems at Carnegie Mellon University&#039;s Heinz College of Information Systems and Public Policy, who coauthored the study. &quot;Contacting customers&#039; friends, a strategy that we call socially based proactive churn management, can help firms reduce churn even in markets without explicit network externalities. Such externalities arise, for example, when prices depend on the firm that friends subscribe to.&quot; </p>      <p>The researchers worked with a major telecommunications provider to design an approach that tested whether contacting customers&#039; friends helped reduce the customers&#039; churn rates. First, the researchers used data from detailed call records to create a model that predicted which customers were most likely to cancel their subscriptions with the company. Then, monthly for eight months, they selected a random group of those customers to be called by trained employees in the firm&#039;s call center, and also randomly selected whether the customers&#039; friends would also be called. The goal of the calls was to identify likely churners and offer them deals to retain them.</p>      <p>The study found that customers likely to cancel their subscriptions who were called but whose friends were not called reduced their likelihood of churning by 1.9% from a baseline of 17.2%. But customers likely to cancel their subscriptions who were called and whose friends were also called reduced their likelihood of churning by an additional 1.3%.</p>   <p> &quot;Our results show that likely churners received a signal from their friends that reduced their propensity to churn,&quot; said Ferreira. &quot;It appears that consumers made decisions based on their conversations with friends about their contracts--in particular, about how much they pay and about the best deals available in the market.&quot;</p>    <p>&quot;If a company knows which customer is planning to leave and who that customer&#039;s friends are, it can work through these friends to reduce the likelihood that the customer will leave,&quot; he adds. &quot;In the end, this increases profits for the company.&quot;</p>  <p>###</p>   <p>Summarized from <em>Marketing Science</em>, Target the Ego or Target the Group: Evidence from a Randomized Experiment in Proactive Churn Management by Godinho de Matos, M, Ferreira, P., and Belo, R. Copyright 2018.</p>   <strong><p>About Carnegie Mellon University&#039;s Heinz College</p></strong>  <p>The Heinz College of Information Systems and Public Policy is home to two internationally recognized graduate-level institutions at Carnegie Mellon University: the School of Information Systems and Management and the School of Public Policy and Management. This unique colocation combined with its expertise in analytics set Heinz College apart in the areas of cybersecurity, health care, the future of work, smart cities, and arts &amp; entertainment. In 2016, INFORMS named Heinz College the number one academic program for Analytics Education. For more information, please visit http://www.heinz.cmu.edu.  </p>                               <p><strong>Media Contact</strong></p>    <p>Caitlin Kizielewicz<br/>[email protected]<br/> @CMUScience     

http://www.cmu.edu

Comments