Telco Customer Retention: an intersection of Network and Business big data

A competitive advantage for Telcos powered by Incelligent’s software framework.

Telco Customer RetentionBackground

Telecom operators are famous for their high spending on marketing, particularly on promotions and other aggressive below the line actions. Many of them use state of the art technologies in order to efficiently segment the market, accurately spot the high yield candidate customers and formulate their offering so as to maximize customer acquisition for each dollar spent.

In parallel, they heavily invest on customer experience and customer retention, as a means to overcome the commoditization of their products & services, and the versatile loyalty of their customers.

The first area of above activities focuses on attracting subscribers while the latter to keep the leak (i.e. customer churn) as low as possible, so that a telco’s business remains profitable.

The challenge

At today’s mostly saturated telecom markets, where penetration with base services has been more or less accomplished, Telcos are struggling for the (re)acquisition and retention of the same pool of customers. Therefore, the two areas of activities merge into a cycle perceived by the customers as essentially the main part of the overall brand equity of the telco.

Therefore, modern times require modern approaches. In essence, TelCos are in a need of a joint view of promotions and customer experience management.

Let’s face it: each promotion that is designed to attract customers has an impact to existing ones. For example, free service credits to attract new sales, up - sales and cross – sales, may frustrate existing, high – spending customers that Telcos need to retain, but also, create side effects by temporarily / ”locally” overloading the network. Moreover they negatively impact the customer perception regarding the value of the Telco VAP. Clearly a promotion has to trade-off between high customer acquisition expectancy and churn mitigation.

But how are these two dots connected? How can one foresee the side - effects of promotional activities on existing high value customers? How to tune such actions before executing them? Who can be wise enough to address such a challenge?

It goes further than big data analytics, data mining or traditional business intelligence. It requires a combination of business expertise and advanced tools that can address non – linear problems exposed to the “butterfly effect”.

An Incelligent solution

Puzzle Complete Feb2016To address this challenge Incelligent uses network, BSS (Business Support Systems), marketing and other data together, so as to reveal the hidden dynamics. After processing such a data set, Incelligent’s machine – learning – based predictive engine (IncelliAna) acquires the full grasp of your challenge, i.e. your network, customers, policies and competitor behaviors. It understands how one factor impacts the other. It comes into position to predict the key performance indicators that are seemingly uncorrelated to your action. This knowledge, extracted from the data, enables Incelligent’s software framework to answer to questions as the following:

  • What will be the impact of a promotion action under consideration to the current subscriber base?
  • How many customers are expected to experience performance degradation further than an acceptable level?
  • At which customer segments will they belong?
  • What will be the impact to their churn probability indices?
  • Which elements of the promotion need to be reconsidered?

With such a view of the future in hands, Telco expert teams come into the position to optimize. The network is no longer a black box managed by some techies at the other floor (or building). Neither are customer behaviors so unpredictable. Internal and external actions are clearly correlated to the business goals and the different aspects of the overall offering of the Telco.

Incelligent opens up a new window for improvement enabling proactive optimization of customer facing actions, network services and big data monetization.

© Incelligent, 2016