How AI can drive transformation in telecoms

Insights

• Integrating AI across various telecom sectors presents numerous opportunities for enhancing customer interactions, optimizing operations, and driving strategic growth.
• Telecom companies can benefit from evolving basic chatbots into more sophisticated AI assistants that offer near-human conversational abilities to enrich the customer engagement process.
• AI's ability to analyze extensive customer data allows telecoms to refine personalization and deliver targeted marketing campaigns. By identifying customer preferences and patterns, companies can create more effective sales strategies and improve customer acquisition and retention.
• AI technologies can enhance network optimization, predictive maintenance, and resource allocation. This can lead to increased efficiency in network planning and operations, ultimately minimizing downtime and improving service quality.

Artificial intelligence (AI) is transforming the telecom industry: From chip manufacturers and telecom equipment manufacturers to cloud providers and software companies, stakeholders across the telecom value chain are tapping into AI's opportunities.

However, adoption remains in its early stages, with many companies still only experimenting with a limited set of established use cases, such as enhancing customer service or generating marketing content, rather than embracing a comprehensive AI-first strategy. Telecoms need to move on to the next stage of their AI journey by choosing the right use cases and strategically integrating AI across a range of activities in their businesses.

Customer engagement across digital and retail channels

The telecom retail industry, covering both digital and physical channels, generates significant revenue, making a strong user experience crucial for customer retention. So it is not surprising that a recent study highlights improved customer experience as a top industry trend for telecom companies. AI-driven customer service enhancements can help telecoms move away from conventional methods that often lead to poor customer service.

• AI assistants: In digital channels, chatbots have been the first point of contact in the telecom industry for a few years now. However, these chatbots, which are pre-scripted, often leave customers wanting more human interaction due to their limited ability to help.
  With generative AI, basic chatbots can evolve into more capable AI assistants that mimic human conversation. This model is also being explored in telecom retail, where robots can help customers by answering queries and guiding them to products.
  Telstra is piloting such AI-driven robot helpers in both digital and physical locations, while Vodafone’s virtual agent, TOBi, handles over 45 million customer conversations each month. These applications both save money and offer quicker and more accurate responses to customers.
• Agentic AI: With agentic AI, the technology is further refined. With generative AI to converse with customers, machine learning (ML) to analyze data and patterns, and reinforcement techniques to make the AI agents learn from actions, the industry is hoping to mimic near-human interactions with customers.
• AI-assisted customer service agents: Accessing customer history is crucial for anticipating queries and providing meaningful interactions. However, support agents often struggle to obtain this information quickly, leading to long wait times and customers having to repeat information across different platforms. AI can help address these challenges in several ways:
  • Quick summaries and sentiment analysis: Instant summaries of customer history and concerns provided to customer service agents will enable faster and more effective query resolution. T-Mobile analyzes previous interactions and alerts agents to customer concerns for more empathetic responses.
  • Omnichannel experience: Customers want cohesive interactions across all platforms to avoid frustrating, disjointed experiences. AI can gather customer interactions from various channels, aggregate the data, and provide it to the customer service agent or intelligent system interacting with the customer, ensuring an optimal omnichannel experience.
    Vodafone reports that utilizing AI for call center activities has improved its net promoter score (NPS), a key metric to gauge customer satisfaction and loyalty.
  • Intelligent call routing: This connects customers to the most suitable agents based on their needs, history, and region, and should reduce the wait time and avoid frustrating misdirected calls.

While these capabilities significantly reduce wait times, ease the work of human agents, and minimize the hassle for customers, the key here is to train customer service executives to effectively query the AI platform for the best results.

AI to drive sales and marketing

Gains from AI in automating tasks, delivering insights and recommendations, and improving personalization are not exclusive to telecoms. However, telecoms are particularly well-equipped to get the best out of AI, thanks to the extensive customer data they gather.

Personalization and targeted campaigns: With AI and data analytics, telecom companies can refine personalization. Deploying AI to analyze customer data — including interactions, preferences, and past purchases — to identify patterns, segment more meaningfully, and uncover insights helps sales teams tailor their pitches and marketing teams plan targeted campaigns. Having AI monitor, analyze, and report social media feedback, competitor market, and customer interactions helps companies adjust product pricing, policies, and marketing efforts to better suit customer needs.

Vodafone’s AI engine matches customers with the right data plan or suggests an upgrade when they are running low on data.

• AI for operational efficiency: AI can handle functions such as data entry, lead generation, reminders, quotation preparation, and appointment scheduling, creating time for humans to concentrate on more strategic efforts that drive revenue.
  Telecom companies like T-Mobile and Vodafone have adopted Salesforce’s intelligent CRM for telecoms, which helps them use their data for lead generation, onboarding processes, scheduling agent appointments, and organize agents by availability and skill for efficient appointment scheduling.
• Request for proposal (RFP) automation: Using an AI model trained on past requests for proposals and responses can streamline the process by generating automated replies. Teams can quickly review and edit these responses, speeding up turnaround times while ensuring quality.

Revenue intelligence

AI-driven revenue management systems analyze data to identify inefficiencies that could lead to revenue loss while also suggesting pricing models, promotions, and strategies to optimize revenue streams.

• Revenue forecasting: AI models can predict revenue trends based on historical data, market conditions, and customer behavior.
• Predictive analytics: Telecom companies have a distinct advantage when it comes to training data. Years of collected customer data, when cleaned and consolidated, can help predict churn and identify high-value customers.

Transforming telecom finance with AI: Generative AI can be integrated into finance teams’ existing software systems that manage processes like accounts payable and receivable, budgeting, and financial closing. This helps with decision-making and uncovering valuable insights from large datasets, including:

• AI-powered financial forecasting and planning: AI leverages predictive analytics to analyze historical data and identify trends, allowing finance teams to generate accurate forecasts that automatically adjust to new information. This leads to more relevant and timely financial planning.
• Improved regulatory compliance: AI, especially through natural language processing (NLP), can automate financial monitoring and reporting, alleviating the resource burden on finance teams. By extracting compliance information from texts, contracts, and policies, NLP enhances accuracy and streamlines the compliance process.
• Optimized cash flow management: AI facilitates real-time cash flow forecasting by aggregating data from various sources, enabling finance teams to make proactive liquidity decisions. This allows for better management of working capital and informed financial strategies, such as taking advantage of early payment discounts or reassessing loan positions.

AI-driven human resource solutions

AI can support HR functions, including:

• Talent analytics: The telecom industry is facing a significant talent crunch, struggling to find experienced engineers while the demand for AI experts and data scientists with domain knowledge continues to rise. In this challenging landscape, leveraging AI to sift through candidate profiles and streamline screening processes to reduce manual workloads becomes essential for alleviating recruitment difficulties.
• Administrative tasks: From sending apt replies to employees or candidate queries to helping them solve routine queries, AI can relieve HR teams from routine tasks that take up much of their time.
• Trends and reports: HR can add more value by leveraging AI and analytics to identify trends, predict turnover, and address skill gaps. AI also analyzes employee feedback and sentiment to measure job satisfaction and diversity, helping HR make informed decisions. It streamlines tasks, saves time, and reduces manual work.

Orange Business is using AI to help streamline recruitment, enhance employee onboarding through chatbots, and personalize training programs based on individual skills.

Network operations

Telecom companies are both increasingly complex and providers of increasingly vital services from consumer applications such as video streaming to enterprise requirements, including remote work and access to cloud infrastructure. AI can support telecoms in providing a reliable service and moving from reactive management to proactive network optimization.

• Network planning: AI and ML are used in network planning activities, including capacity analysis, infrastructure design, and traffic management. Predictive maintenance uses AI-driven simulations to identify potential network issues before they occur, allowing operators to plan and implement solutions in advance. Optimizing resource allocation by analyzing traffic patterns and predicting future demands ensures that the network can scale appropriately to handle increased data traffic.
• Network field operations: Network field operations in telecom involve managing extensive infrastructures, including multiple sites and diverse equipment. AI aids in scheduling maintenance and dispatching technicians based on both real-time and historical data. Generative AI-driven insights now enable technicians to quickly and accurately diagnose problems by querying the model.
• Predictive maintenance: Predictive maintenance is the backbone of a reliable telecom network. ML algorithms monitor and analyze network data to anticipate problems and issue alerts. Technicians can use generative AI to query the system for details about the problem and possible solutions. While there is still a human in the loop, predictive maintenance is far easier with the help of AI.

Deutsche Telekom's T-Car intelligently plans 60,000 kilometers of fiber optic routes while also employing AI for cyber defense, enhancing both infrastructure development and security. Meanwhile, China Mobile is aiming to become a pioneer in autonomous networks, which are engineered to operate with minimal human intervention.

AI for the future

Companies such as SK Telecom and Deutsche Telecom are jointly developing telecom-specific large language models (LLMs) to ensure more accurate insights and enhanced network security. As the industry evolves, the future will likely be driven by hybrid models that combine LLMs with smaller, specialized language models tailored to specific tasks, such as network optimization, customer service, and predictive maintenance, meaning telecoms can harness the scalability and versatility of LLMs while benefiting from the precision and efficiency of specialized models designed for particular uses.

In an industry rich with data and focused on improving the bottom line, it's essential for telecoms to adopt an AI-first strategy. Balakrishna D.R. (Bali), executive vice president, global services head, AI and industry verticals, says this about AI adoption by telecoms: “AI is the catalyst for the next generation of telecoms, enabling intelligent networks and insight-driven businesses. Infosys is collaborating with telecom partners to help them go AI-first, which will unlock their full potential, while embracing responsible and ethical AI practices. We are also contributing to the industry and open-source initiatives and fostering a partner ecosystem that not only drives innovation but also takes a strong leadership role in shaping the future of the industry.”

To do this, they need to get their organizations ready for AI, putting a solid data architecture in place, onboard talent, and have an appropriate governance framework. However, telecoms must conduct a cost-benefit analysis to ensure that their AI investments align with their financial goals.

While concerns around data privacy, cybersecurity, the need for skilled AI talent and high energy consumption by AI solutions might cause hesitation, addressing these issues will transform AI from a mere trend into a cornerstone of strategic advantage.