Telecommunications companies, AKA Communication Service Providers (CSPs), oversee a significant amount of network data. This data, collected from network devices, servers, and applications, holds the potential to offer deep insights into network performance and inform strategic decision-making.
Sasa Crnojevic, EMEA AIOps Business Principal for Service Providers at SAS
Data-driven networks represent a significant opportunity for CSPs to integrate network services with business operations seamlessly. By effectively harnessing their network data, CSPs are able to gain a comprehensive understanding of their network performance, leading to more informed and agile business strategies.
However, many CSPs are not fully leveraging this valuable resource, often due to organisational and technological barriers such as data silos, outdated legacy systems, and skills gaps. Overcoming these issues is key to CSPs being able to transform their data into actionable business decisions.
Integration of data driven networks with business operations
The integration of network services with business operations can pave the way for enhanced customer experiences, optimised operational efficiency, and aid with the identification of new revenue streams. For instance, by analysing data on network usage and performance at a central location or ‘on the Edge’ directly from a device, CSPs can identify and address issues proactively, plan future network and new capacity rollouts, and so ensure a more reliable and satisfying service for their customers.
Data-driven insights can help CSPs optimise network performance, reduce downtime, as well as plan network and manage network resources more effectively, leading to cost savings, improved service delivery and eventually better ROI. Additionally, with a deep understanding of network usage patterns and customer behaviours, CSPs can develop targeted services and personalised offerings, opening up new avenues for revenue generation — especially for upcoming 5G services targeted to the enterprise segment.
The integration also opens up opportunities for CSPs to enter new markets and forge strategic partnerships. By analysing data from various sources, CSPs can identify emerging trends and potential areas for expansion. For example, they might discover increasing demand for Internet of Things (IoT) solutions in certain industries, prompting them to develop tailored offerings that address specific industry needs, such as 5G private networks for example.
Additionally, data-driven insights can help CSPs identify potential partners, such as technology providers or content creators, to collaborate on innovative solutions and expand their market reach.
Identifying opportunities
The telecommunications industry is at the forefront of embracing transformative technologies such as AI and Generative AI (Gen AI). AI algorithms, including machine learning models and Large Language Models (LLMs), play a pivotal role in analysing and deriving insights from vast datasets.
However, the effectiveness of AI models is contingent upon the quality of data inputs, as otherwise it can lead to Gen AI model ‘hallucinations’. Therefore, ensuring high-quality data acquisition and management practices is essential for optimising AI-driven decision-making processes and enhancing predictive capabilities.
Gen AI represents a significant leap forward in AI technology, offering capabilities to automate operational tasks, enhance creativity, and optimise business processes within CSPs. By leveraging it alongside traditional AI approaches, CSPs can accelerate innovation, as well as drive operational efficiencies, and deliver personalised customer experiences.
However, Gen AI alone does not serve much purpose for CSPs, but when combined with AI-driven algorithms and processes throughout orchestration and governance, it can help them achieve enhanced network automation and optimisation. GenAI enables CSPs to summarise the context of the problem, and it can define prompts for and work as a co-pilot with AI.
For example, where AI detects anomalies or predicts network performance trends,
Gen AI can work with AI to proactively mitigate potential issues, through AI-driven decisioning, before they impact service delivery. This predictive capability not only enhances network reliability but also enables CSPs to optimise resource allocation, reduce operational costs, and maintain competitive advantage through Next Best Action (NBA) in a rapidly evolving marketplace.
The telecommunications industry is also witnessing the emergence of low and no-code data and AI platforms, which are transforming how CSPs approach data-driven strategies. These platforms allow users to build applications and automate processes without requiring extensive technical expertise.
This democratisation of technology is particularly valuable for CSPs, as it helps address the skills gap and accelerates digital transformation, through the MLOps or Models Factory approach. It enables non-technical staff to participate in data analysis and application development, broadening the pool of employees who can contribute to data-driven initiatives.
Low and no-code AI solutions allow for rapid prototyping and deployment of applications, enabling CSPs to respond quickly to market changes and customer needs. By reducing the reliance on specialised IT staff and minimising development time, low and no-code platforms can significantly lower the costs associated with digital transformation.
However, there remains a need for some employees with deeper technical and data science skills to oversee the development and monitoring of more complex AI models and decisions, ensuring that AI is compliant with upcoming regulations and follows a Trustworthy AI programme.
Current challenges
Despite the clear advantages, several challenges hinder CSPs from fully capitalising on their network data. Data is often stored in isolated systems, preventing a holistic view of the network. It means the involvement of data scientists or data engineers is needed every time you want to implement a new use case or change or tune an existing one.
This fragmentation and ‘non business friendly AI systems’ makes it difficult to analyse and derive meaningful insights from the data. Many CSPs rely on outdated legacy systems that are not equipped to handle the volume, variety, and velocity of modern network data. These systems can be inflexible and costly to upgrade or replace.
Transforming raw data into actionable business decisions requires specialised skills in data analytics, machine learning and network management. Many CSPs face a shortage of these skills within their organisations, hindering their ability to leverage data effectively.
Another significant obstacle is the cultural shift required within organisations to embrace data-driven decision-making. This involves fostering a data-centric mindset across all levels of the organisation and encouraging collaboration between different departments. CSPs must invest in training and upskilling their workforce to ensure employees are equipped with the necessary skills to analyse and interpret data effectively.
Overall, data-driven networks offer a transformative opportunity for CSPs to bridge the gap between network services and business operations.
As CSPs continue to evolve, those that embrace data-driven strategies and adjust their processes accordingly will be well-positioned to lead the industry into a new era of connectivity and customer satisfaction. By investing in the right technologies, fostering a data-centric culture, and addressing data privacy and security concerns, CSPs can harness the power of data-driven networks to drive growth and innovation in the telecommunications industry.
Author: Sasa Crnojevic, EMEA AIOps Business Principal for Service Providers at SAS
This article originally appeared in the October 24 magazine issue of IoT Insider.
