Customer Segmentation in Telecom

In the past 2 decades there has been a humongous increase in the accumulation of customer data increasing the need for data mining aimed at customer relationship management (CRM) and understanding the customer. It is a very well known fact that the Telecom industry holds the most structured amount of Customer data due to the nature of revenue generation from calls. With the rapid increase in competition customers pose different risks, making it imperative to implement different treatment strategies to maximize shareholder profit and improve revenue and maintain the customers in order to prevent the churn.

Segmentation is the process of developing meaningful customer groups that are similar based on certain characteristics and the customer behaviors. The goal of segmentation is to know your customer better and group those customers to apply that knowledge to increase profitability, reduce operational cost and also customer service. Segmentation can provide a multidimensional view of the customer for better engagement. An improved understanding of customer risk and behaviors enables more effective portfolio management and the proactive application of targeted solutions to increase profitability

To compete with other providers of mobile telecommunications it is important to know enough about your customers and to know the wants and needs of your customers. To realize this, it is needed to divide customers in segments and to profile the customers. Another key benefit of utilizing the customer profile is making effective marketing strategies. Customer profiling is done by building a customer’s behavior model and estimating its parameters. Customer profiling is a way of applying external data to a population of possible customers. Depending on data available, it can be used to prospect new customers or to recognize existing bad customers. The goal is to predict behavior based on the information we have on each customer. Profiling is performed after doing customer segmentation.

Segmentation is a way to have more targeted communication with the customers. The process of segmentation describes the characteristics of the customer groups (called segments or clusters) within the data. Customer segmentation is a method to segregate customers with common characteristics across the entire current, prospective, target and lost customer segments. These segments can then be treated as distinct groups and analysis can be conducted based on the characteristics showcased by the customer. Customer segmentation analysis can save significant marketing effort since the company can then focus only on the required high profitability customers. Multiple tools and dashboard reports can be used for customer segmentation. Using clustering algorithms one can define segments from the entire customer base into groups identified on the basis of various parameters like

• Location/ Demographics (GDP and economy)
• Product/ Services
• Line of business (B2B/ B2C)
• Volume of business/ Company size (Corporate customers)
• Third party distributors/ Vendors/ Alliances
• Customer spending capability
• Customer billing trends
• Age and profiles
• Gender
• Socioeconomic (Culture, ethnicity, education, income)
• Economic conditions (Spend Analysis, Price sensitivity)
• Values, attitudes and beliefs (Customer loyalty)
• Life cycle (Period of association)
• Knowledge and awareness of other products (Inquiries, Campaigns, Surveys etc )
• Lifestyle/ Personality (Interest, Tastes, Preferences)
• Acquisition Method (How was the customer acquired)
• Credit and payment history (Historical data and credit rating)
• CLV (Customer Lifetime Value)

Accurate market segmentation is essential in order to use analytics to successfully acquire customers. Using analytics in making the right marketing and operational decisions can result in getting a greater market share of customers and marketing products with higher returns. Focusing on the right segment predicted using BI tools will not only result in higher customer lifetime value but also increase customer satisfaction via customer loyalty. Customer Relationship Management (CRM) and Business Analytics and Optimization are inseparable.

Customer Profiling

Customer profiling provides a basis for marketers to ’communicate’ with existing customers in order to offer them better services and retaining them. This is done by assembling collected information on the customer such as demographic and personal data. Customer profiling is also used to prospect new customers using external sources, such as demographic data purchased from various sources. This data is used to find a relation with the customer segmentation that were constructed before. This makes it possible to estimate for each profile (the combination of demographic and personal information) the related segment and visa versa. More directly, for each profile, an estimation of the usage behavior can be obtained.

Predictive Segmentation and Analytics

Executives, senior management and sales operations teams in telecom companies generally follow traditional segmentation methods that generally do not give them a deep understanding of which parameters in the telecom product or service offer actually drive customer behavior and define their segmentation. Traditional approaches to segmentation are strictly experiential, looking at past customer choices and behavior. Predictive segmentation, on the contrary looks futuristic, examining how customers falling in the same customer segment may respond to changes in the duration, offering, channels, pricing and competition. The predictive approach uses the market’s response to these future incidents, in combination with past behaviors, as the basis for segmenting. It also gives the sales team a high degree of confidence in their decisions about what to do differently for each segment, because potential moves have been “simulated.” Analytical tools based on predictive segmentation help in the telecom companies living the future with a near real realization of the results and outcomes.

This forward-looking perspective comes from putting representative customers from the  target markets through a simulated buying exercise, presenting them with fully described offerings from all the relevant suppliers in the marketplace. Elements of the offerings are varied using a specific sampling approach called experimental design, which allows a marketer to project the impact of many stimuli by testing just a few of them. Using mathematical formulas to select and test a subset of combination of variables that represent the complexity of all the original variables, marketers can model hundreds or even thousands of stimuli accurately and efficiently.

Predictive segmentation provides a tried and tested formulation for designing, testing, measuring returns and rolling out sales and marketing operations. However the segmentation can not be very rigid and exhaustive since it is human behavior and thus in total the consumer behavior can not be woven into the BA tool as artificial intelligence. Understanding of all expected customer behaviors can not and should not be built into the segmentation solution as it might be resulting into biased decision making. A most common mistake generally done by market operations team is to use segmentation and marketing analytics with the usage of a single format segmentation or a single predictive model. 

SPSS defines the five predictive imperatives to maximize customer value with predictive analytics.

• Base your customer strategy on predictive profiles
• Predict the best way to win the right customers
• Predict the best way to grow customer relationships
• Predict the best way to keep the right customers longer
• Use predictive intelligence at every customer touch-point

Benefits of using predictive customer segmentation

The predictive analytical approach not only produces forecast for predicted segments; it also gives users a high degree of confidence to the executive management on their decisions about what to do differently for each segment. Any potential market scenarios and moves have been “pre-tested” and were embedded while building the segment definitions. These departures from past approaches make predictive segmentation ideal for the highly competitive and fast changing telecommunication sector. Historic data even though it is good in format and extensively detailed can not necessarily be a good predictor of future behavior. This approach of predictive customer segmentation thus typically tests and predicts different permutations and combination of forward-looking moves and thus showcases see the potential impact in a variety of simulated moved. This insight allows executives to choose which business plans to proceed ahead with and which plan to discard, by segment. Thus predictive customer segmentation can help the CEOs and COOs of Telecom companies to Redefine Strategy for their Future

 



*** More detailed report is present separately