Telecoms service providers are betting on managed services to drive growth as competition intensifies and data margins are squeezed, says GlobalData, but while forecasts indicate a managed network services CAGR of 8,6% between 2025 and 2030, rising risk emerges as AI-driven automation threatens margin-eroding commoditisation.
Consequently, service providers must develop economically viable managed services that integrate tokenomics, the research group says.
Gary Barton, research director, Enterprise Technology & Services at GlobalData, comments: “While an 8,6% CAGR trails the growth rates generated by technologies like AI and cloud, managed services for enterprise clients present a compelling opportunity during a period when telcos are struggling to achieve sustained growth. AI automation can counteract margin erosion and help extend managed services to new market segments like SMBs.”
However, automation is a double-edged sword: while it can make managed services easier and cheaper to deliver, there is a risk that enterprises will expect costs to fall and standards to improve. There is also the problem that while automated services are perceived as low-cost, the investment needed to build models and the ongoing token consumption once agents and LLMs are up and running presents a challenge to service providers.
“Telcos should be aware that tokenomics has the potential to break managed services if it is not handled properly,” Barton continues. “Service providers need to factor in the cost of AI from the beginning and not lose sight of it. It is always the case that certain features will become table stakes as technologies mature, but swallowing the cost of token consumption will become a poison to telcos if they do not manage it correctly and carefully.”
Despite the potential pitfalls, AI-powered managed services do offer new opportunities to service providers and can help them to monetise existing network infrastructure and the data derived from operating very large networks serving large numbers of enterprise customers. However, these solutions must also be built with trust at their core.
“Enterprises are already struggling to monitor and control IT environments and ensure business continuity – helping them to do so will offer value to providers,” says Barton. “Providing observability with agentic AI agents which can identify and fix problems or suggest remediation to humans is the start, and service providers can do this because of their vast network experience.
“This can then be extended into areas such as AI agent management, AI-enhanced network security, and even AI consumption management,” Barton adds. “However, providers will need to design these solutions with trust at their core as the most effective solutions will need to access sensitive data from enterprise’s network and IT estates.”