Pro-Active Care

Customer-centric predictive models

Zhilabs Pro-Active Care solutions predict customers’ behavior based on Artificial Intelligence and Machine Learning models applied to real-time network and customers insights.

Prediction of Customer Complaints

Nowadays with the fierce telco market competition to attract new customers, Operators need to move from reactive to proactive Customer Care solutions if they want to retain their subscribers. Anticipate to subscribers’ issues and identify proactively those ones who are thinking of contacting to the call center, can make the difference between remain or leave.

Zhilabs ML models predict the subscribers likely to complaint by integrating and correlating a set of key data sources (CEM, usage patterns, KPIs, KQIs, tickets, NPS surveys) what allows Operators to anticipate and take effective corrective actions to avoid customer complaints and therefore reduce churn and increase NPS.

Customer Experience Models & NPS Alignment

Evaluate customers’ satisfaction is a complex task for Operators since not all their customers are willing to answer NPS surveys and in many times these surveys don’t reflect the reality.

By applying Augmented Analytics, Zhilabs can predict NPS metrics for 100% of Operator’s subscribers base from a subset of NPS surveys.

Zhilabs solution cross-correlate customer experience models, network data and NPS surveys to forecast customers’ satisfaction level for the entire network.

Use NPS predictions as a driver for network self-management provides great benefits like:

  • Guarantee customer loyalty and increase the likelihood of gaining new ones.
  • Forecast your business growth, cash-flow, as well as assess the health of your network and overall
  • Automate the network management and optimization based on customer’s overall relationship with your business.
  • Reduce customer complaints and Churn.
  • Identify behavioral trends and tracking business performance over time.

Churn Propensity Models

Since keeping a customer is usually cheaper than the cost to get a new one, in highly competitive markets where subscriber growth is limited, providing the highest Quality of Experience is a must to differentiate from the competition.

Being able to identify likely churners may help the operator in keeping the #1 position, especially if they are high-value subscribers (high ARPU)

In order to face this challenge, Zhilabs provides Machine Learning models to compute list of possible churners, allowing Operators to trigger tailor-made campaigns to improve end-user satisfaction.

Clustering of Subscribers

Network inventories are difficult to maintain, and in most occasions contain inconsistent information. Additionally, Operators need to understand similar elements in the network (in the RAN, in the packet core, in IMS) depending on different variables (e.g. customer types, ROI/ARPU, etc.)

Zhilabs’ Pro-Active Care solutions automatically identify all data sources and provide an accurate view of the network and its services by clustering together elements that are similar, by different business variables (ARPU, Heavy Users, VIP, Enterprises,) or technical parameters (e.g. daily profiles, traffic profiles), helping Operators to automatically segment their customers’ base.