Baku, January 14, Tamilla Mammadova, AZERTAC

In the era of digital transformation, telecommunications are no longer just a means of connecting people—they are evolving into a complex ecosystem of data and services. Today, the speed, quality, and personalization of services directly depend on how effectively operators can collect, analyze, and leverage network and customer information. Artificial intelligence (AI) and big data analytics are no longer a luxury but a necessity, opening new opportunities to improve customer service, reduce costs, and create innovative products. For Azerbaijan, this is particularly relevant: the telecom market here is not yet burdened with outdated systems, which creates favorable conditions for implementing modern AI solutions and digital transformation. This allows for rapid improvement in service quality and the ability to offer customers new, personalized products.

The Azerbaijan State News Agency (AZERTAC) presents an interview with Stanislav Streltsov, a recognized expert in data and AI, specializing in strategy and digital transformation.

– What key trends in data management and AI in telecommunications do you consider most applicable to Azerbaijan?

- The first and probably the most important trend is predictive network maintenance. Telecom operators use AI to continuously monitor the condition of network equipment. AI helps forecast potential failures before they occur, allowing for proactive maintenance and reducing the risk of unexpected disruptions for customers. This technology has already proven effective and is widely used by telecom operators worldwide because it truly saves costs and improves service quality.

Another key trend is customer churn prediction. It is essential to analyze user behavior, including how customers use services, how often they contact support, and which services they activate or deactivate. Based on this data, AI can identify customers at high risk of leaving and provide the company with an opportunity to engage them proactively, offering personalized solutions or incentives to retain them.

Automation of customer service through chatbots and AI assistants is another significant trend. Such systems enable clients to receive faster responses to routine inquiries while reducing the workload on call centers. This leads to faster and more convenient service for customers, while operators can significantly reduce support costs.

These technologies are especially promising for Azerbaijan because the market is not yet overloaded with outdated or overly complex systems, unlike more mature telecom markets. This simplifies the adoption of modern solutions. However, for these technologies to work effectively, investment in basic data infrastructure is crucial to ensure that data is high-quality and readily available for analysis.

– What are the main benefits AI provides to telecom operators and their customers?

- For telecom operators, AI brings several key advantages. It allows significant reductions in operational costs through the automation of routine processes, including network maintenance, handling standard customer requests, and identifying problems. AI also increases network efficiency by enabling faster detection and resolution of disruptions, better load distribution, and improved service quality. Additionally, AI helps improve customer retention, as it allows for more accurate predictions of user behavior and timely delivery of personalized offers, reducing the likelihood of churn.

For customers, the benefits are also tangible. They receive faster resolution of issues, as standard requests can be handled automatically. Customers are offered more relevant plans and promotions that align with their needs and habits. Furthermore, predictive network maintenance improves connection quality, reducing service interruptions and signal problems. It is important to note that the success of AI initiatives depends heavily on the quality of implementation. Poor data preparation, analytical mistakes, or misaligned use cases often result in AI projects failing to deliver their promised benefits and cause disappointment.

– How can telecom companies in Azerbaijan use data to enhance service quality and create new products?

- Telecom operators can leverage the data they collect in several ways. For service quality improvement, network data allows the identification of areas with weak coverage, overloaded segments, and recurring problems. This enables operators to make more informed infrastructure investments, enhancing connection quality and increasing customer satisfaction.

Regarding new product development, there are opportunities, especially in the B2B segment. Telecom operators can offer IoT connectivity for industrial companies, provide location analytics for retail, or create fleet management solutions. These applications are closely related to the core telecom business and have real commercial value. Attempting to enter entirely new business areas often encounters challenges, primarily due to a lack of industry expertise outside telecommunications. Therefore, it is more practical to start with products adjacent to the core business rather than immediately pursuing completely new and complex domains.

– What steps are necessary to build effective data management infrastructure and develop AI?

- The first step is establishing a data governance system, which involves defining data ownership, quality standards, and access rules. This is primarily an organizational task, as even the most advanced technology cannot function effectively without a clear structure and accountability. The next step is creating a centralized data infrastructure, consolidating all data in one place so it can be analyzed and used for AI projects. Depending on the company’s structure and specific use cases, this may involve a data lake, data warehouse, or data mesh architecture.

Equally important is the computing infrastructure. Companies must decide whether to use on-premises GPU servers or cloud services, basing this choice on workload analysis and cost considerations to avoid unnecessary spending. Skilled personnel are another essential factor. Specialists who understand both AI and the telecom business are in high demand, and training existing employees is often more realistic than trying to hire rare and expensive talent. Finally, it is important to start with small, high-value projects. This approach allows companies to quickly achieve results, gain expertise, and build internal trust in new technologies.

– What are the priorities in data security and privacy?

- Data security and privacy require careful management. First, strict access control is essential, as most data breaches occur internally rather than from external attacks. Encryption of data both at rest and in transit is a standard practice and a fundamental requirement for modern systems. Data minimization is also important: only data that is genuinely necessary for business purposes should be collected and stored to reduce risk. Regular security audits and penetration testing help identify vulnerabilities early and prevent potential threats. A key challenge is balancing security and operational efficiency, as absolute protection can slow processes and increase costs, so finding a reasonable compromise is essential.

– Which AI and data initiatives are most promising for Azerbaijan specifically?

- The most promising initiatives include network optimization and predictive maintenance, as these solutions have proven effectiveness, clear return on investment, and are relatively easy to implement. Automation of customer service, such as chatbots for routine requests and automated problem diagnostics, also shows strong potential. Fraud detection is another area with clear business justification and measurable results, demonstrating rapid economic benefits once implemented. At the same time, it is advisable to be cautious with complex new business models, research projects without clear commercial applications, or blindly following all AI trends.

– What global AI trends do you see, and how should Azerbaijan prepare for them?

- Globally, foundation models such as GPT are becoming increasingly accessible, reducing the need to build models from scratch and allowing for faster integration of AI into business processes. Edge computing, where data is processed on devices rather than solely in the cloud, is gaining importance due to low-latency requirements. Regulatory oversight around AI is also increasing, as exemplified by the European AI Act, which signals that companies must be prepared for new rules and compliance requirements.

Preparation for these trends should involve investing in basic infrastructure, including reliable connectivity and modern networks. Establishing partnerships with cloud providers ensures flexibility and scalability. Developing education and training programs helps build a workforce ready to work with AI. Additionally, maintaining a clear and balanced regulatory environment is crucial, one that supports business and innovation without imposing excessive restrictions.

It is important to remember that implementing AI is slower and more complex than marketing materials often suggest. Successful projects are built not on chasing every new technology, but on careful execution of proven solutions that deliver clear business value.