Mobile Number or Hand-phone number has come to represent an element of person’s identity and hence there is an implicit inertia that is built into a person’s action while mobile numbers are to be changed. Thus, mobile companies were building up an embedded loyalty into customers’ behaviour, when mobile numbers portability was not in force – there was a reluctance to change the operator even when the customer was not satisfied with the services or charges or plans. With the advent of “Mobile Number Portability” (MNP) this very paradigm of “forced loyalty” was broken – as the customer could shift the operators without changing the mobile number. This resulted in companies becoming more customer friendly in terms of their plans, pricing, and service.
Customer Loyalty – Its impact on the Cost
A survey of telecom literature indicates that the cost of acquiring a new customer is at least 5 times the cost of retaining an existing customer. With mobile telephony penetration curve flattening, market players are eyeing the same market, which is fast shrinking. Mobile connectivity becoming the order of the day irrespective of the geographical reach – be it either urban or rural – customers are forcing network providers to dynamic plans which are pocket-friendly to the customer and at the same time to invest in the back-end infrastructure to reduce call-drops.
All these market dynamism on the players’ part is expected to play a role in creating the “hygiene factor” for customer retention. However, the players are depended on the “invested” customers to increase their use of data services (2G / 3G) and value-added services (VAS) to increase their top-line. Thus, it becomes all the more imperative for the market players to retain their existing customers and induce them to build their relationship for a higher mobile spend on various services offered by them.
Customer Retention Programs:
The new mobile market paradigm has forced CRM Managers to shift gears in designing their retention programs to become more dynamic and customer focused. The success of these programs will be measured on the ‘Δ’ it creates in most singular metric on which most telecom companies are measured – ARPU (average revenue per user).
For even the best of the retention programs to be successful, it has to be able and amply supported by the back-end analytics infrastructure that maximizes the impact of the program by aiding the decision makers to zoom in on the right target segments for each customized retention program. However, Retention/CRM managers find that they get less than acceptable returns on their retention programs. This is mostly because these programs are not targeted sharply enough. A large proportion of customers targeted are often the ones who would not have churned in the 1st place.
Churn Management – Analytics at play
Predictive Analytics is one of the key tools deployed to bring sharpness to the customer targeting exercise, and to optimize the marketing spend, in a scientific data-driven manner. Analytics is used in churn management – in the form of regression models, decision trees – with an aim to identify the potential attrite, and flag their behaviour with a goal to reduce the total churn. Analytics at work, for churn management, is built with an aim of knowing in advance “Who will churn?”
For optimal targeting, i.e. allocating resources on basis of probability of attrition (maybe along with profitability of the customer or other factors), basic data variable/attribute requirements are:
– Demographic data from customer information file like age, sex, zip code etc.
– Contractual data from service account file such as pricing plan, activation data, contract identification etc.
– Usage & Payment data from billing system such as the number of calls, airtime, fixed line time, the total amount spent, number of calls made to customer care centre, change in price plan etc.
Churn Management – Current Scenario
However, in the process of building a single customer view, handling many variables sometimes causes dilution while focusing on price elasticity aspect of telecom services – essentially, the focus on time series data on customer’s payments and calls made. Looking at different price-quantity coordinates and drawing insights from the change in trends/patterns is an important exercise which often gets overlooked.
Time to the expiry of the contract is an important variable diligently tracked by the telecom operators. Out of attriting customers, majority attrite after their contract expires. So, sometimes because of this reasoning, Survival analysis, which tries to answer both questions “Who will churn?” as well as “When will he/she churn?” gets ignored.
The Key Question:
And of course, none of these methods answers the question “Why does a customer churn/attrite?”
And the answer to this question calls for a rigorous and cordial marriage between telecom CRM strategies and telecom CRM analytics.
Try deep learning using MATLAB