The telecom minister has set targets for BSNL, especially on service quality. What steps are being taken to achieve them?
Quality of service (QoS) is our primary operational focus. We have put in place a mechanism on network optimisation and monitoring. We have deployed dedicated QoS teams in every one of our circles, conducting regular test drives and KPI (key performance indicator) reviews. Quality parameters on calls, including rates of call drops, are monitored continuously and are within the acceptable benchmarks of the regulator. We are committed to continuous improvement in network optimisation. We track customer-reported issues and have structured complaint analysis circlewise to identify and address geographic gaps in coverage and service quality. We launched in January pan-Indian VoWiFi, which improved indoor coverage and call clarity in weak signal areas. We have reached a meaningful quality baseline, and the gap with private operators is closing with each successive optimisation cycle.
On tariffs BSNL is undercutting all carriers. To what extent has that helped in customer retention, especially when your 4G network is still expanding?
Let me be precise on this. Our strategy is not tariff-led undercutting in the traditional sense. Our approach is built on two sequential pillars; first, deliver a quality network that the customer trusts and finds reliable and, second, customer service and satisfaction. Once this is established, it will grow our commercial base and average revenue per user.
What is the timeline for setting up the next 23,000 towers? Have purchase orders been issued?
Yes, this is firmly in motion. We are discussing with vendors. Advance purchase orders were already out but the final ones require some incremental changes. It is in the works.
What is the timeline for completing the 4G rollout?
We’ve rolled out most of the towers. Now it’s only optimisation, and we’ll keep on adding to it. The network is built entirely on indigenous technology developed by Tata Consultancy Services (TCS), the Centre for Development of Telematics, and Tejas Networks, and is among a handful of domestically developed telecom stacks in the world. Now, we’re continuously optimising, adding coverage in white-spot areas and progressively upgrading 3G subscribers to 4G.
When will the 5G initial rollouts begin?
For commercial 5G mobile service, we are targeting a soft launch in Delhi in six months. Our vendor has started work on the rollouts. We will start from here. The advantage BSNL has is that all our deployed 4G radio equipment is 5G-upgradable through a software upgrade and minor hardware additions, meaning the transition to 5G does not require greenfield towers.
How is BSNL looking at integrating and implementing AI?
AI is a strategic priority for us and we are pursuing it across three distinct domains. One is network intelligence. TCS has embedded a Cognitive Network Management System in our 4G infrastructure. This system leverages AI and machine learning for real-time anomaly detection, predictive maintenance, and automated fault resolution. Alongside this, we are deploying Self-Organising Network (SON) capabilities, which enable our network to self-configure, self-optimise, and self-heal, significantly improving call quality and reducing manual intervention. Both these capabilities have a direct impact on quality of service.
Next is customer experience. We have launched AI-powered digital assistant “Vaani”, enabling 24×7 conversational support for service queries, complaint registration, and bill payment.
We have also deployed an AI-based Plan Recommender and a recharge expert system, which analyses a customer’s usage profile and recommends the most suitable plan with minimal inputs. These tools are improving customer satisfaction scores.
The third is Rural Digital Infrastructure, with Edge AI on BharatNet. We are evaluating a model where gram panchayats act as micro edge computing nodes, distributing data processing closer to the user rather than routing everything to central data centres.
Paired with a remote fibre monitoring system for rural backhaul, and a federated learning architecture that allows AI models to train on distributed data without centralising sensitive user information, this could transform BSNL’s rural network from a passive connectivity pipe into an active digital infrastructure platform. This aligns with our objectives on national data sovereignty and BharatNet’s broader Digital India mandate.