--- title: "5G Adoption Tops 30%, but Revenue Growth Still Lags" type: "News" locale: "en" url: "https://longbridge.com/en/news/281999971.md" description: "5G adoption has surpassed 30% globally, with nearly 2.8 billion connections. However, revenue growth remains sluggish, primarily due to uneven monetization and reliance on consumer revenues. Only 20% of operators have deployed standalone networks, critical for advanced services. The industry is shifting towards AI-driven revenue models, with a focus on open architectures and network slicing to enhance monetization. Despite improvements in user experience, the challenge lies in unlocking value from existing 5G infrastructure. Telecom operators are also prioritizing cost optimization through AI to manage rising expenses." datetime: "2026-04-08T08:10:55.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/281999971.md) - [en](https://longbridge.com/en/news/281999971.md) - [zh-HK](https://longbridge.com/zh-HK/news/281999971.md) --- # 5G Adoption Tops 30%, but Revenue Growth Still Lags As 5G accounts for more than 30% of global mobile connections, the telecom industry is entering a make-or-break phase in which future investments and demand for next-generation network and compute infrastructure will hinge on whether operators can finally monetize standalone networks, AI-driven services, and enterprise use cases. Despite the rapid uptake, monetization remains uneven. According to the Global System for Mobile Communications Association’s (GSMA) Global Mobile Trends 2026 report, 5G accounts for nearly 2.8 billion connections worldwide, yet mobile revenue growth has been “sluggish for much of the 5G era,” often barely keeping pace with inflation. The report states that only around 20% of operators with 5G networks have deployed a standalone (SA) architecture, considered critical to unlocking advanced services such as network slicing and low-latency applications. In the absence of scalable enterprise use cases, operators continue to rely on consumer revenues, which contribute about 70% of total earnings, limiting the upside from premium 5G services. While 2025 showed early signs of improvement through pricing changes and evolving service strategies, the industry remains in the early stages of translating 5G scale into sustainable returns. EE Times spoke to Arunav Roy, senior VP and business unit head of connectivity at Cyient, who said, “Over the last four decades, there has been a technology shift every eight to 10 years, whether it is 2G to 3G, 3G to 4G, or 4G to 5G. Each time, the investments were monetized, and there was a compelling need and experience change for the consumer.” However, that cycle has slowed. “For the first time, after the rollout of 5G, we have crossed that eight to 10-year boundary, but we are not yet moving to 6G,” he explained. “Monetization of 5G assets is still at about 30 to 40%.” Recent data also points to a gradual shift in how operators approach monetization through AI. According to GSMA Intelligence, 25-50% of telco AI deployments launched in the six months leading to December 2025 were focused on driving revenue, up from 10% to 15% a year earlier. The focus in 2026 is expected to shift toward experimenting with revenue models for telco AI capabilities. Breakdown by region of AI deployments, showing the share focused on internal efficiency gains vs. external product revenue. (Source: GSMA Intelligence) The gap between investment and returns is shaping the industry’s next phase. Roy said that while user experiences have improved, they have not fundamentally changed. “You still have similar use cases today, perhaps slightly faster, but not at 10× or 20× improvement,” he said. “The question is whether we are able to unlock value that allows monetization of the existing 5G infrastructure.” The push to unlock value is driving a broader shift in how telecom networks are designed, deployed, and operated. Roy said traditional telecom networks have largely operated as closed systems, even in multi-vendor environments. “The first major shift we are seeing is toward openness,” he said. “Open architectures allow operators to enable network slicing and move toward offering “network as a product,” particularly for enterprise use cases. “That allows them to target enterprise use cases and improve monetization,” he added. “It impacts the entire value chain.” Telecom networks are increasingly evolving in a manner similar to mobile devices, in which hardware, operating systems, and applications have become decoupled over time. Initiatives such as the Open Radio Access Network (O-RAN) are enabling external developers to build applications on top of network infrastructure. This has led to the emergence of RApps and XApps, which support functions such as configuration management, network assurance, remote diagnostics, and energy efficiency. Roy said the market for these applications is expected to reach around $700 million in the coming year, with Cyient itself developing about 25 such applications for deployment across global telecom operators. The move toward open and programmable networks also supports broader connectivity goals. Roy pointed to the role of satellite communications in extending coverage, citing an example of a European operator increasing network reach from 83% to 98% through satellite integration. “The goal is to extend coverage; once coverage is available, enterprises need to operate seamlessly,” he said, adding that private wireless networks and network slicing will play a key role in enabling such use cases. Telecom operators face pressure to manage rising costs. Roy said heavy investments in network infrastructure have made cost optimization a priority alongside monetization. AI is emerging as a key tool to address this. “AI enables greater autonomy, reduction in operating expenses, fewer physical interventions for maintenance, more remote diagnostics, better assurance, less configuration drift, and improved energy efficiency,” he said. This shift toward AI-led optimization is also influencing how network infrastructure is designed and deployed. Roy said telecom operators are adopting a heterogeneous compute approach, using GPUs for parallel processing and AI workloads, particularly at the edge, while continuing to rely on CPUs for control plane and latency-sensitive tasks. “It is not about choosing one over the other, but about selecting what fits the use case,” he said, pointing to trade-offs between cost, energy efficiency, and performance. At the network level, this transition is part of a broader move toward intelligent and automated operations. Roy described a three-layer model adopted at Cyient for intelligent network modernization: network engineering, autonomous networks, and cognitive operations. The company operates network operations centers across multiple regions, where engineers work alongside AI agents rather than being replaced by them. The approach combines domain knowledge, vendor ecosystem expertise, experience in solving use cases, and proprietary AI tools, including Cyient’s Bizmon AI platform. While cost optimization remains the immediate priority, these capabilities are also enabling new revenue opportunities. Beyond operational efficiency, new enterprise use cases are beginning to take shape. In manufacturing, digital twins allow remote monitoring and control of factories, reducing the need for physical intervention. In logistics and retail, image sensing and high-definition video analytics support automated sorting and tracking of parcels. “These use cases can be implemented using private networks, which are more reliable and secure,” Roy said. Market dynamics for 5G adoption vary widely. While India is one of the largest telecom markets by users, it lags in revenue generation. In an earlier conversation with EE Times, former GSMA chairman and board member Mike Short said, “India is probably the second largest market by customer numbers, but it’s also about the fifth largest by revenue. Low value means you need a lower cost base to support a low value market.” This gap between scale and revenue makes India a larger opportunity. Roy said Cyient’s earlier focus was on fiber buildouts, and 4G and 5G transitions in other regions, but initiatives such as BharatNet and the growth of autonomous networks are shifting attention toward the Indian market. Work on next-generation networks has already begun, but Roy said 6G remains a few years away. “It is still four to five years away from making an impact, likely around 2029 to 2030,” he said. In the meantime, telecom operators are focusing on improving returns from existing 5G investments. Future developments could include the use of terahertz spectrum, AI-native interfaces, and integrated sensing and communication, with standards bodies such as the 3rd Generation Partnership Project (3GPP) and the International Telecommunication Union (ITU) playing a key role in shaping the roadmap. At the same time, the industry itself is evolving beyond traditional services. “Services remain a foundational element, but they are increasingly supported by IP, platforms, and solutions,” Roy said, pointing to a shift toward more integrated and outcome-driven business models. The market opportunity reflects this shift. Roy estimated the network engineering services market at $500 billion to $600 billion, with an additional $100 billion to $150 billion opportunity in autonomous and cognitive network operations. As operators look to close the gap between 5G investment and returns, Roy said success will depend on how effectively they combine openness, automation, and new enterprise-driven use cases to create sustainable revenue streams. For an industry that has built 5G at scale, the next phase will depend less on deployment and more on demonstrating that it can pay for itself. What Makes mmWave Practical for India’s 5G Engineering MWC 2026 Concludes as Telcos Pivot to AI Arm Positions Neoverse for AI and Telco Networks at MWC Yashasvini Razdan is the Senior India Correspondent for EE Times. With a bachelor’s degree in electronic engineering from Mumbai University and a post-graduate diploma in business journalism, she has served in various senior editorial roles in leading Indian electronics and business publications, where she reported on semiconductor innovations, deep tech startups and the Indian electronics ecosystem. She is passionate about combining her technical know-how and journalistic acumen to chronicle the evolution of India’s electronics ecosystem for a global audience. You must Register or Login to post a comment. This site uses Akismet to reduce spam. Learn how your comment data is processed. ### Related Stocks - [XTL.US](https://longbridge.com/en/quote/XTL.US.md) - [515050.CN](https://longbridge.com/en/quote/515050.CN.md) - [KNCT.US](https://longbridge.com/en/quote/KNCT.US.md) - [XLC.US](https://longbridge.com/en/quote/XLC.US.md) - [NXTG.US](https://longbridge.com/en/quote/NXTG.US.md) - [VOX.US](https://longbridge.com/en/quote/VOX.US.md) - [159695.CN](https://longbridge.com/en/quote/159695.CN.md) - [159994.CN](https://longbridge.com/en/quote/159994.CN.md) - [515880.CN](https://longbridge.com/en/quote/515880.CN.md) - [159583.CN](https://longbridge.com/en/quote/159583.CN.md) - [IYZ.US](https://longbridge.com/en/quote/IYZ.US.md) - [IXP.US](https://longbridge.com/en/quote/IXP.US.md) ## Related News & Research - [ANALYSIS-AI debt sales reshape global corporate bond markets](https://longbridge.com/en/news/288215628.md) - [Ellevest Inc. Has $9.16 Million Stake in Netflix, Inc. $NFLX](https://longbridge.com/en/news/288237740.md) - [Sfmg LLC Has $3.01 Million Stock Holdings in Palantir Technologies Inc. $PLTR](https://longbridge.com/en/news/288169364.md) - [Norges Bank Takes $942.54 Million Position in Motorola Solutions, Inc. $MSI](https://longbridge.com/en/news/288168704.md) - [Amazon.com, Inc. $AMZN Shares Purchased by Cheviot Value Management LLC](https://longbridge.com/en/news/288231201.md)