Analyzing Customer Retention
Earlier this week, the Market Research Group of the Atlanta Chapter of the American Marketing Association heard two different perspectives on "Utilizing Customer Analytics Towards Increasing Customer Retention and Corporate Profitability." Both presenters discussed how to use analytics to better understand the composition of your customer base in terms of loyalty, satisfaction, share of wallet and profitability.
One presenter nicely laid out the traditional customer lifecycle as acquisition, growth, retention and win-back, focusing on the retention part of the customer lifecycle for his presentation. From a research perspective, another way to think about it is:
Prospective Customer/Competitive Customer
New Customer or Early Life Customers
Current Customers
Customer Retention
Lost Customers (Customer Churn)
Customer retention and lost customer research differ by the type of respondent. Churn research is traditionally performed exclusively with former (lost) customers while customer retention research is performed with former and/or current customers. While lost customer research focuses on why they left, customer retention focuses on why they might be considering leaving and what might make them stay. Otherwise, the two surveys focus on the same thing – a better understanding of what can be done to increase customer satisfaction and retention rates. As one presenter correctly pointed out, performing marketing research to measure current customer satisfaction, likelihood for retention and reasons for customer loss is only one part of retention analytics process.
When a company realizes it is losing customers at an unacceptable rate, and that keeping customers is up to 10 times less expensive compared with acquiring new ones, they have begun the first phase of the process, which is recognition of a potential problem. The next phase is leveraging historical customer activity data and creating customer profiles to identify the specific points along the customer lifecycle continuum that customers are leaving or canceling. Once the key continuum points are identified, more in-depth research and analysis can be conducted to pinpoint the specific drivers of customer churn at those key continuum points.
For those of you that have attempted retention analysis, you know that the above can be the easy part, for the next phase is to take specific and measured actions at each of the continuum points based upon findings from the analysis. For example, a trigger can be set for all customers fitting profile A that reach the 90th day of doing business with your company – perhaps it is a reminder postcard, special offer or personalized phone call. Setting up these triggers and executing a sales and marketing driven strategy based upon multiple trigger points can be tricky, time consuming and expensive. For that reason, it is critical that you know the true lifetime value of customers within each of your established profiles so that you can perform a cost benefit analysis and decide to what degree you are willing to take proactive retention measures.
In completing the cycle, it is then critical to measure the success of your program in real terms, such as customer retention/profitability and performance changes on the key churn drivers as measured by your marketing research tracking program. Due to the comprehensive nature of retention analytics programs such as these, time and money costs are high, as is the need for very senior level corporate involvement across multiple divisions/functions. Depending upon many variables within your business and industry, it may be time to consider a retention analytics program -- large scale or small.
Cell Phone Survey Regulations
As all of you savvy marketers and market researchers know, it's getting harder and harder to reach respondents and keep response rates up in the United States. This is true of all research methodologies. One of the key factors considered to be contributing to the issue is how busy average Americans keeps themselves, leaving little time for taking surveys or other more leisurely activities. Other factors are the various acts of government legislation, such as the Federal Trade Commissions' (FCC) Do-Not-Call list, various state anti-spam laws as well as the Federal Can-Spam Act, and the more recent Federal Communications Commission regulations regarding reaching cell phone users via automated dialing, a technique used for the overwhelming majority of consumer surveys.
In an order released on September 21, 2004, the FCC adopted a fifteen day safe harbor for calls made using an automatic telephone dialing system (autodialed) or prerecorded message to a recently ported (landline number switched to a cell phone) wireless phone number in violation of the Telephone Consumer Protection Act of 1991 ("TCPA"). The Commission arrived at its decision by balancing the desire of the public to be free from unwanted calls with the technical limitations on callers' abilities to continually update call lists to remove ported numbers. The FCC felt that providing companies with this safe harbor would not infringe upon consumers' privacy rights because of its limited duration and consumers' ability to prevent all unsolicited calls through the National DNC Registry.
What does this mean to researchers? It does force all of us to spend the additional time and money bumping our lists up against known cell phone lists for scrubbing, which can cost an extra several cents per record. There are also issues with representative samples, as an estimated 2.5 percent of Americans have only cell phones with no traditional landline service (there are 169 million cell phones in usage in the United States). With time, it will become more and more important to represent that increasing portion of the population through weighting or other methods. As with list scrubbing, making sure sample populations are representative will only add more cost and time to projects, however minimal. The challenges can be overcome, but it is becoming that much more important to plan accordingly during the design and strategy phase of research projects. |