As telcos move to cloud-native architectures, they have the advantage of rich telemetry, providing valuable insights for optimising network performance, enhancing security and improving customer experience. There is a caveat: the complexity can lead to monitoring tool sprawl, alert fatigue and spiralling costs. Can observability platforms utilise this data and turn it into something useful?
From wholesalers to altnets, the telecoms industry is adopting cloud-native architectures to drive down operating budgets, improve customer experience and launch compelling new services.
Cloud-native promises agility and scalability, but the new architectures are increasingly disaggregated, dynamic and distributed. This makes managing networks, customer experience and compliance far more difficult. The surge in telemetry data that needs to be processed and understood can overwhelm teams and obscure basic metrics required for performance and reliability.
“Traditionally, telcos relied on OEM vendors to provide observability tools bundled with their network equipment. This approach has led to fragmented visibility, data silos and observability blind spots,” says Gaurav Gupta, the strategy advisor at Cisco Splunk. “So most telcos today are still primarily focused on metrics and logs [providing] visibility into device health, network utilisation, alarms and basic performance KPIs. This approach is increasingly seen as insufficient in the face of NFV and cloud-native architectures.”
Observability provides the deep end-to-end visibility needed to track microservices, containerised workloads and hybrid networks in real-time. Without this 360-degree view, telcos risk losing control of performance, customer experience and compliance.
“Containerisation fundamentally changed the game and spelled the beginning of the end for traditional monitoring methods,” says Bill Hineline, the field CTO at Chronosphere, a developer of observability for microservices and containers. “The sheer scale of telemetry from a containerised environment demands performant and scalable observability.”
Traditional monitoring can only offer isolated snapshots of what is happening in different parts of the network. It lacks the context to know how systems are interconnected and how one failure impacts another. An observability strategy can cut through the noise, turning data into real-time intelligence so telcos can lead on performance and protect revenue by reducing churn while meeting compliance demands.
Observability helps telcos work through the flood of telemetry data by filtering and contextualising signals into intelligence ready for smart decision-making. This enables proactive issue detection, faster resolution and much stronger SLA compliance. It also paves the way for AIOps and autonomous networks. These are critical to maintaining competitiveness now and in the future.
Disparate observability adoption
While many leading telcos are talking about Observability – including Vodafone, T-Mobile, Telefonica, Rakuten, BT, Orange Telenor and TalkTalk – take-up remains inconsistent. “Most telcos have a goal for full-stack observability,” explains Timo Jokiaho, the CTO of telco solutions at SUSE. “But they start with a more modest path and are applying a phased approach.”
A 2024 study by New Relic found that 69% of firms in tech and telecoms had at least four tools for observability, and 23% had over eight tools. A third spent over US$1 million a year on observability tools. The study found telcos are “spending time and money tool-hopping to understand the different aspects of their business and avoid costly outages.” The flipside is that significant outages cost a lot of money – a quarter of telcos said that outages of 1 hour cost over US$1 million, adding up to a median annual cost of US$12 million. So even with observability tool sprawl, many respondents felt they were achieving an annual ROI.
Filtering signal from noise
Cloud-native and 5G environments generate vast amounts of telemetry data, risking data overload. “Regardless of the industry, the challenge remains the same: too much data, too little clarity,” says Chronosphere’s Hineline. Teams struggle with the signal-to-noise ratio, unclear ownership, and the cultural shift required to transition from monitoring to modern observability. The trend I see gaining traction across the board is moving from ‘collect everything’ to ‘collect what matters’.”
Filtering out data closer to the source and making a stronger correlation between different systems within a network will help reduce the noise. “Observability must extend beyond simply surfacing data. It needs to connect the dots to become truly valuable,” adds Hineline.
Metrics do matter
Observability is rapidly becoming more than an operational necessity for telcos. It can be a competitive advantage that enables them to understand and adapt their infrastructures in real time.
Prianca Ravichander, the CMO at BSS developer Tecnotree, emphasises the competitive dimension: “Observability is evolving into a powerful competitive differentiator. It allows operators to understand their business as a living, adaptive system. And those who master the translation from telemetry to value will command the next era of telecom leadership.”
Stewart Baines