Customer Value Analytics and Acquisition in Telecoms
Synopsis
A leading regional telecoms company aimed to expand its customer base, but it faced considerable obstacles, primarily in growing its retail sales channel. Moreover, in the competitive landscape of customer acquisition, the task of identifying and attracting high-value clients proved to be quite challenging. Traditional approaches frequently depended on broad campaigns, which could lead to inefficiency and the risk of alienating existing customers.
Furthermore, legal restrictions on approaching competitors’ customers hampered targeted and effective customer acquisition efforts. In light of these challenges, Trusted Data employed advanced machine learning techniques to uncover a method to engage with existing customers and support organic subscriber growth.
Client Overview
🔸Modern regional mobile operator with +$3Bn revenues and a sizeable consumer base.
🔸c.1.5 million in postpaid consumer subscribersÂ
🔸Top 3 in-country market position but new customer acquisition had stalled in the previous 5 quarters.
Client Problem
“By harnessing the power of AI and analytics, our telecommunications client was seeking to pivot and revolutionise their approach to customer targeting whilst remaining compliant with regulatory frameworks. Leveraging customer advanced lifetime value (CLV) metrics, they could identify and empower top-tier customers with the highest propensity to refer new clients. Applying such a strategic approach would drive organic growth and strengthen their community of loyal customers“.
Â
Analytics Delivery
🔸Utilising advanced analytics, we wanted to empower our client to assess the value of its existing customer base and craft targeted messaging to encourage referrals.
🔸Detailed data analysis of customer data including usage patterns, purchase history and engagement levels wth specific contacts in their network
🔸This enabled the company to categorise customers according to their potential value, customer network strength and likelihood to refer friends.
The Challenge
⦿ Regulatory framework in the clients’ business territories prevented direct marketing to off-network users.
⦿ How to identify value and opportunity from the clients’ existing network user data.
⦿ Creating a meaningful and targeted marketing scheme to incentivise referrals from the right existing customers.
⦿ Intensive modelling requirements including customer base exploration, customer network analysis, customer revenue and value analytics
⦿  Establishing a methodology for ML network-based models to identify valuable customers exhibiting a degree of relevance and spending similarity to existing high-value clients.
Trusted Data Approach
Our modelling objective was to develop sophisticated predictive models to identify top customers most likely to act as brand advocates, allowing the company to focus its referral programs effectively.
Personalised messaging strategies could then be developed for these high-value customers, offering tailored incentives and rewards for successful referrals.
To achieve this a data and modelling strategy needed to provide answers to the following:
⦿ Estimate ARPU per customer and designate high value customers in the subscriber network
⦿  Model relational strength matrix of current subscribers with off network prospects
⦿ Validation of models estimating propensity of customers to switch over
⦿ Identify best mechanisms to target existing customers
⦿ Evaluate model performance for a referral and revenue sharing scheme
OUR IMPACT
%