The ‘AI-enhanced digital twins for best NPS network - Phase II’ Catalyst aims to transform the net promoter score (NPS) from a passive metric to a proactive governance system. To achieve that goal it is harnessing AI, digital twins and big data to proactively monitor and manage network issues.
Turning passive customers into active promoters through AI and digital twins
Commercial context
In the ultra-competitive telecoms industry, customer satisfaction is crucial for CSPs. Yet they can struggle to identify the root causes behind customer dissatisfaction. Most CSPs measure their NPS using customer surveys, but this approach can result in fragmented data and delayed insights. In turn, these shortcomings hinder the real-time resolution of critical issues, such as network latency or service gaps.
In practice, siloed feedback across product, network and service domains prevents an holistic approach to issue resolution. At the same time, aggregated and delayed survey results make root-cause analysis difficult. A lack of data on the impact of past interventions can further complicate the allocation of resources to improve NPS.
Solution
The first phase of the ‘AI-enhanced digital twins for best NPS network – Phase II’ Catalyst established a unified data-driven metrics framework for product, network and service domains. However, operational silos remained. For example, customer complaints continued to be isolated from network metrics, leading to sub-optimal solutions.
The second phase of the Catalyst is formalizing NPS governance. It is aiming to replace fragmented, siloed NPS management with a dedicated organizational structure to support closed-loop issue resolution. The project is also deploying genAI to eliminate cross-departmental barriers and empower the governance team with predictive analytics. Natural language chatbots enable non-technical stakeholders to query complex network KPIs in seconds.
Meanwhile, the Catalyst is also using AI to develop digital twins across customer journeys to help streamline NPS optimization. With network data spanning the full user journey, the CSP will be better placed to proactively resolve root causes.
“By utilizing network data and employing AI4NET Large-Scale Localized Statistical Channel Modeling technology, we modeled PCI and AAU beam scenario – planning to optimize network configuration and minimize interference,” explains Frank Xie, Deputy Head, Radio Network at China Mobile Hong Kong, one of the champions of the Catalyst. “The digital twin will model real-world networks, preemptively resolving interference and parameter conflicts for seamless performance and superior service delivery.”
“We were able to analyze diverse data sources, identify problematic cells and make adjustments to optimize performance, all within an automated network environment,” adds Frank Xie. “The project used innovative technologies such as high-fidelity digital twin simulation, a multi-modal parameter optimization agent, an intelligent beam optimization algorithm, and a realistic network simulation algorithm to significant reduce weak coverage areas, decreased interference levels, and enhanced signal quality, savings in new site deployment costs.”
The project is drawing on numerous TM Forum assets. Those assets include the Telco NPS Management Framework and the Business Process Framework. Additionally, it draws on the Autonomous Networks Business Requirements and Framework and the Digital Twin – Network Optimization API Profile.
Application
Pilot deployments of the Catalyst solution have reduced churn, enhanced network reliability and improved retention rates. “As a result, this validates NPS as a strategic growth driver rather than mere KPI,” says Başar Günyel, Senior Manager of Network Quality at Vodafone Türkiye, another champion of the Catalyst. “For example, the new governance framework enabled an improvement of 26% in the NPS score – a result achieved by aligning network, marketing and customer service teams with unified metrics.”
“Our customers’ happiness and experience are our main-focus areas, and we measure this through complaints, surveys, and other feedback mechanisms. NPS surveys provide valuable insights into customer sentiment, and we leverage various network data sources to correlate this feedback with actual network performance.”
Having prioritized NPS as a core KPI, Vodafone Türkiye is now leveraging TM Forum’s assets to shape its organizational culture and develop a human-centric NPS methodology. As well as sustaining this customer experience improvement, the telco is looking to scale its solution across Vodafone’s operating companies.
Wider value
The Catalyst team claims its solution is the first industry-wide implementation of agent-based collaboration between network, operations and customer service teams, using shared data pipelines and more than 28 restoration algorithms. The project is also pioneering the use of vertical AI to broaden access to data. As a result, it bridges the ‘digital divide’ between technical and non-technical teams. CSPs from across the globe are supporting the project. As well as China Mobile and Vodafone Türkiye, Advanced Info Service (AIS), Entel Perú, Globe Telecom, PT Telekomunikasi Selular and Saudi Telecom Company are champions of the Catalyst.
“By establishing governance frameworks, democratizing data access, enabling cross-domain collaboration, and driving precision marketing, the Catalyst redefines NPS management as a proactive, cross-functional discipline, setting a new paradigm for telco customer experience innovation," concludes Başar Günyel.