Reader's Poll

Which of the following technologies/concepts are likely to witness significant traction this year?
Any data to show


Tele Data

Mobile Subscribers Yearwise comparision

Controlling Churn: Operators turn to predictive analytics to reduce attrition

March 22, 2017

Controlling Churn: Operators turn to pre...

By Suhale Kapoor, Co-founder and Executive Vice-President, Absolutdata Analytics

India has the second largest telecom market and the third highest number of internet users in the world. In the past one decade or so, the Indian telecom sector has registered strong growth to reach a subscriber base of 1,058.86 million as of March 2016. Despite this growth however, the industry has failed to halt the customer churn. In the past two years, there has been a substantial increase in churn, with the prepaid segment bearing a bigger brunt than the post-paid segment.

The country’s telecom market faces stiff competition with the presence of more than 10 mobile operators, predominantly in the prepaid segment. The burgeoning rate of customer churn is one of the biggest challenges for telecom companies, compounded of late by increased saturation in terms of players and offerings. About 96 per cent of all mobile subscribers are constantly switching between mobile service providers to avail of incrementally lower prices. The monthly churn rate in India is approximately 6 per cent. According to researchers and industry observers, some of the major reasons for the high churn rate are the availability of lucrative alternatives, the ease of switching to other brands and a substantial growth in prepaid mobile subscribers.

Reliance Jio Infocomm Limited’s (RJIL) entry into the market with compelling offers such as free voice calls, cheaper roaming rates and inexpensive data plans has led smaller mobile service providers to either exit the fray or face high churn. Every service provider is today coming up with innovative tariff plans and trying to redesign them with better offerings to lure customers. Bigger service providers are coming up with attractive promotional schemes and packages to retain prepaid customers. But these have been proved ineffective as the churn rate is extremely high. Customer retention, therefore, is becoming critical for maintaining a stable customer base.

In this scenario, telecom companies are realising that analytics can be a saviour in tackling churn in the long run. Predictive analytics can not only help in identifying and predicting customers who are likely to churn, but can also help incentivise and retain them by sending relevant messages and offers. By harnessing the available data effectively, telecom companies can derive incremental benefits such as reduced churn, increased customer base and take real-time fact-based decisions that can help increase the return on investment significantly. In order to control churn, telecom companies should be mindful of the following:

Customer lifetime value analytics

Customer lifetime value (CLTV) analysis is the best way to prioritise marketing efforts. It attempts to predict the value that a customer brings to a business over the entire journey and sends relevant offers for retaining the customer. Although telecom companies follow this to a certain extent, it is important to assign a quantifiable monetary value to each customer in order to prioritise various sets of customers. The lifetime value model provides the predicted yield from each customer over the customer lifecycle.

CLTV analysis can help:

• Target customers who are likely to churn.

• Identify customers with high usage and propose upgraded plans to suit their needs.

• Develop a targeted marketing plan for each customer segment based on churn likelihood.

• Recommend services that a customer would be interested in buying based on their past behaviour. High priority customers can be given offers/packages with higher discounts, preferential treatment through personalised services, etc.

• Identify accounts that might require in-person sales efforts to grow and nurture.

Predictive analytics

Predictive analytics techniques such as churn prediction modelling, customer retention analytics and next best offer recommendation can help better align the customer segmenting/targeting exercise and optimise marketing communication strategies for reducing the level of customer churn. Predictive analytics is being widely used by telecom companies as it not only measures the churn propensity of a customer but also suggests a retention strategy. Churn modelling is an easy analytics option with relatively low complexity, immediate results and wide acceptance among functions. Thus, many companies deploy it to tackle the churn issue.

The techniques used in analysis such as regression models and decision trees help identify the potential attritions and flag their behaviour with a goal to reduce the total churn. These modelling techniques study the past behaviour of customers and identify profitable customers who can be retained through personalised actions. The value of the suggested offers can be tailored to suit customer profitability.

Campaign management and analytics

Creating marketing campaigns that are highly targeted or hyper-personalised can make a significant dent on the churn cycle. Hyper-personalised campaigns help increase response, adoption, conversion, etc. by identifying key segments and personalising services/plans as per customer requirements, thereby fighting churn directly. Also, by designing powerful awareness programmes through road shows, innovative mass promotion with various schemes on data and voice services, the campaigns can help in customer retention.

An integral part of campaign management is campaign analytics as it helps measure the campaign performance and identify which messages/offers yield better incremental results. Campaign analytics done in a phased manner can help accurately track campaigns that had an impact in controlling churn or increasing customer acquisition. The framework to measure the effectiveness of a campaign usually comprises the following:

Pre and post design

• Observe sales/response in the pre-period.

• Apply the marketing stimulu.s

• Observe sales in the post-period.

• Compute promotional and post promotional benefit.

• Pre-, promo- and post-period adjustment to compute incremental benefits.

Test and control

• Assign customers to test and control groups.

• Send the new marketing stimulus to the test group.

• The delta in test and control groups is attributed to the marketing stimulus.

• A profile comparison of test vs control.

Cross-selling and up-selling

A commonly used solution to handle churn in any industry where the customer base is varied and huge is through cross-selling and up-selling services. A common issue faced by many telecom service providers is how to increase yield from the current subscribers, or how to improve ARPUs. The solution is supported by analytics techniques using association rules and transaction histories of existing customers. Analytics-driven cross-selling and up-selling campaigns provide higher returns. Moving beyond financials, they also increase retention and reduce the number of customers who might churn in the near future. These models may be integrated with real-time decision engines and developed as real-time cross-selling and up-selling systems. They are usually planned by the marketing department and executed through in-bound customer-facing teams, mainly call centres.

Optimised pricing strategy

Developing a pricing strategy that is abreast with the current market scenario is the key to surviving in a volatile market condition. With RJIL entering the market with highly subsidised prices, both big and small players have to come up with plans/packages, be it voice or data, that will help them sustain the pressure. Pricing analytics backed by market research can play a significant role here as it can help  chalk out an optimal pricing plan for a particular package. The analytics techniques used here also take into account competition and macroeconomic factors, thus being cautious about the current market scenario and delivering what best suits the current and potential customers.

To sum it up, telecom companies have to depend heavily on analytics for real-time decision making, and predictive analytics can significantly manage and reduce customer churn. The key benefits that telecom companies reap from predictive analytics are diverse, ranging from identifying new opportunities and managing risk to reducing churn and improving efficiencies.


To post comments, kindly login

 Your cart is empty


OFC conferenc

Monday morning