Customer Experience Management

With the coming of the mobile internet era, mobile services and products have become increasingly rich; market competition has become more homogenous; and customers have become more discerning with their service experience. Honing in on customers and their quality of experience (QoE) has become the only way for operators succeed in the market.

To accurately determine and manage QoE, leading mobile operators have developed customer experience management (CEM). According to the TeleManagement Forum, CEM is subordinate to QoE management. In CEM, the emphasis is on measuring user experience directly in real time and defining a user experience index. CEM is used to determine the service experience of certain users or groups in the hope of finding and solving problems before complaints are made. In this way, customer satisfaction can be improved.

A CEM system can compile key user indexes and convert them into information about user experience. This helps an operator recognize and understand user experience, and business departments can optimize existing resources accordingly. The system can provide data for fault analysis so that a fault can be located quickly. It can also observe, assess, and improve in real time the quality of user services and add value to network operation. By analyzing user and service information contained in the data, operators can better understand user behavior and service development. This helps with business planning and expansion. The CEM system mainly consists of QoE assessment system, data source collection and transformation, and mass data computing platform.

QoE Assessment System

QoE can indicate network and service experience in almost a quantitative way, and it can show the gap between the actual and expected quality of service and network. Many standards organizations have researched QoE and proposed their own measurement criteria. The top-down CEI-KQI-KPI modeling approach is widely used in the industry:

●    KPI is a key performance indicator for a network or network element.

●    KQI is a key quality indicator for different services. It is closely related to user experience and can be divided into service accessibility, retainability, and integrity.

●    CEI is a customer experience indicator. It provides an objective measurement of customer experience.

A modeler needs to collect the original data from a network, develop a formula for calculating KPIs, and design models for mapping KQI to KPI and mapping CEI to KQI. Non-technical factors also need to be quantified when building a QoE assessment system. The weight of CEI and KQI in the calculation formula should reflect the degree of customer care, and the ultimate goal of the system design is to improve customer satisfaction and value. To build a QoE assessment model that accurately reflects customer experience, the modeler should have a deep understanding of services and be experienced in network optimization.

Data Source Collection and Transformation

The CEM system needs to support data collection from various heterogeneous data sources in order to accurately reflect the quality of end-to-end services and customer experience. Data sources include:

●    customer data. This includes detailed customer information such as name, service subscription, and SLA.

●    active probing. This includes drive test and dialing test. A test pack is added to the existing network to obtain end-to-end metric data.

●    passive probing. This is used when the network or equipment supplied by a third party does not automatically provide data for service analysis. The data is obtained by monitoring the network in a way that does not involve adding a test pack to the network.

●    NMS data. This includes network KPIs and network element KPIs, both of which are obtained through the EMS or NMS northbound interface.

●    xDR data. This includes MR, CDR and TDR data, which are delivered through relative interfaces on the equipment.

●    network asset management. This provides detailed configuration data for all network elements.

The CEM system also needs to extract and transform these heterogeneous data sources for statistical analysis of different dimensions.

Mass Data Computing Platform

A mass data computing platform is essential for effective customer experience management. The platform has the following main functions:

●    mass data collection, extraction, and storage

●    real-time aggregation computation and monitoring of QoE indexes 

●    data mining for user behavior and QoS analysis

●    separation of computation and data based on the network type and service product in order to ensure system expandability

●    unified interface management system that provides user interfaces for real-time QoE monitoring, specialized analysis reports, fault handling, and SLA management.

ZTE’s self-developed user behavior analysis system, called ZXUN UBAS, can provide all these functions. With embedded scripting ability, it can customize extract, transform, and load (ETL) processes. It can also compute and monitor in real time any index from any data source on any dimension. It stores and analyzes mass data in the data center and supports online report definition and query. A unified web portal provides abundant UI interaction. Depending on the scale of the data, centralized or distributed networking can be implemented. ZXUN UBAS has been applied successfully in many markets worldwide.

Conclusion