Enhancing survey-based net promoter scores with a data-driven approach is long overdue

Commercial context

Like any other business, CSPs face the critical challenge of maintaining customer satisfaction and loyalty. However, CSPs also encounter unique challenges: global mobile penetration is plateauing, and intense competition for value-added services leaves little room for revenue growth. This is expected to become still more severe over the next five years. Yet CSPs have a significant advantage: their advanced network technology and breakthroughs in other areas, such as AI, enable them to gather more intelligent and accurate customer feedback.

This puts them in a prime position to transition away from historical methods they have relied on for their net promoter score (NPS) - this is used to differentiate themselves and provide a foundation for improving customer satisfaction and loyalty. Traditionally, NPS management has depended on surveys, which often suffer from data distortion and small sample sizes, ultimately limiting their effectiveness. The advance of new technologies in NPS management is therefore essential to overcoming these limitations and enhancing customer experiences.

The solution

The Data-to-NPS: Boosting NPS using decision intelligence Catalyst project, which won a 2024 Catalyst Award in the Outstanding Catalyst - Attendees' Choice Award category, is developing an innovative solution to enhance NPS management. This initiative is driven by a dedicated team of experts from various fields, tasked with enhancing the traditional survey-led approach with a proactive, data-driven methodology. The project is being undertaken by multiple industry-leading CSPs and communication technology service providers, including Telkomsel, China Mobile (Guangxi, Zhejiang, Shaanxi), Globe, AIS, STC, Huawei, Eastcom Software, and Sudo Technology.

The project breaks customer satisfaction and NPS management into three distinct modules: product, network, and service. By using advanced technologies such as digital twins and AI, the abstract concept of customer satisfaction and NPS can be represented in digital form, ultimately establishing a management system that is ‘visible, manageable, and optimizable’. The team has benefited from the use of TM Forum DT4DI Reference Architecture (IG1310A), which ensures continuous contributions to the DT4DI Top Use Cases (IG1310C). This provides the means to develop solutions that are scalable and can be efficiently implemented by CSPs worldwide.

Product NPS

The intelligent marketing decision-making framework helps to identify potential dissatisfied customers from multiple dimensions, such as data and voice plan overages. The most suitable products can then be recommended through the most accessible channels to the most needy customers, resulting in highly personalized matches between customers and products. This is supported by marketing assistant Copilot for planning and personalized script generation, significantly improving efficiency and success rates. A large language model enhances the accuracy of real-time data and maintains the integrity of data assets.

Network NPS

A spatio-temporal digital twin engine and traffic autonomous zones are also used to correlate NPS with network issues. The digital twin system significantly expands the pool of satisfaction samples, improving the reliability of NPS measurements. Its high-performance data processing engine enables real-time data analysis, automatically verifying and correcting anomalies.

Additionally, the modelling algorithm, which incorporates both long and short period analysis, is employed to address the challenge of discordance between customers' subjective perceptions and objective network issues. The short period is used for user experience modelling, while the long period focuses on the correlation analysis between the user's short-period experience and their long-period derogatory behaviour, also considering the impact of the forgetting curve. This approach enables identification of potential detractors across the entire user network.

The spatio-temporal digital twin engine helps pinpoint potential detractors and their corresponding timeslots. This enables comprehensive analysis of network performance, identifying root causes of dissatisfaction, including precise locations and times. This comprehensive approach provides more precise identification of potential network detractors, pinpointing the root causes of dissatisfaction, such as the location and time of poor quality or unusual experiences.

Service NPS

The service NPS is an intelligent system that integrates data from multiple service-related systems. It identifies potentially dissatisfied customers and the reasons for their dissatisfaction, such as long queues at branches, through four key dimensions: physical channels, electronic channels, the 10086 hotline, and outbound marketing. It automatically generates task orders for the corresponding outlets, enabling swift responses and effective tracking. Additionally, user accessibility has been modeled to ensure optimal allocation of service resources, which guarantees feedback is translated into actions for service improvement.

Wider application and value

The data-driven approach to NPS management provides substantial benefits for businesses working across the telecommunications industry and wider digital services market. The increased accuracy and reliability of inputs and subsequent NPS measurements enable CSPs to promptly address network issues impacting NPS. This improves their own customer satisfaction and loyalty goals, but also those of enterprise customers who rely on high-quality, reliable networks to ensure their own customer loyalty targets. Measurable impacts determined by the project team include a potential 20% improvement in NPS scores, a 15% reduction in churn rates, and a 2% increase in revenue.

This sets a new benchmark for NPS management and demonstrates the effectiveness of data-driven strategies, paving the way for other industries to adopt similar methods, and raising the overall standard of customer experience management. More broadly, the solution enhances the quality of digital services, which are increasingly essential in daily life. By improving NPS, CSPs can reduce churn and customer complaints, leading to cost savings that can be passed on to consumers. According to Indra Mardiatna, CTO, Telkomsel Indonesia, “TM Forum DT4DI & Huawei helps unlock the mystery of detractors’ unspoken truth.”