Telecom’s Margin Reset: Why Operators Need AI-Enabled Cost Control, Not Just More Traffic
Telecommunications operators are entering a margin reset. Demand for connectivity continues to rise, but the economics of the sector remain pressured by flat-to-down ARPU, capital intensity, pricing competition, customer churn, legacy systems, and increasingly complex service expectations. More traffic is not enough if the operating model cannot convert that traffic into better returns.
PwC’s 2026 telecom outlook frames the challenge directly: global traffic is soaring while ARPU is flat to down and revenue growth remains modest, making AI-enabled cost control and efficiency essential. McKinsey’s 2026 work on telco AI adds a caution: most operators are piloting AI, but only a small share report sizable balance-sheet impact. The issue is not whether AI can help telecom. The issue is whether leaders can redesign the operating model enough to capture the value.
AI-enabled cost control will only improve telecom margins when it is embedded into operating routines: customer care, field service, network maintenance, provisioning, billing, churn management, capital allocation, and service assurance managed through clear ownership and measurable performance.
The Traffic-Revenue Gap Is Structural
Telecommunications has long faced a difficult economic pattern: usage rises, but revenue does not rise at the same pace. PwC projects global telecom service revenue to increase from US$1.15 trillion in 2024 to roughly US$1.32 trillion in 2029, a modest 2.8% CAGR. At the same time, mobile ARPU is expected to tick down slightly, and fixed broadband ARPU remains constrained by the decline of legacy fixed voice revenue.
This creates a structural margin challenge. Operators must continue investing in fiber, 5G, network resiliency, cybersecurity, customer systems, and service quality, but incremental revenue per unit of traffic remains under pressure. If cost structure and operating complexity do not change, AI-driven demand can intensify the margin problem rather than solve it.
The leadership implication is clear: the margin reset cannot be handled as a cost-cutting cycle alone. It requires a more disciplined operating model that reduces waste, improves service quality, automates repeatable work, shuts down legacy complexity, and ensures that capital and labor are deployed against the work that creates value.
AI Is a Margin Tool Only When It Changes Work
AI has become a central theme in telecom strategy, but the practical value depends on whether it changes how work gets done. McKinsey reported in 2026 that virtually all surveyed telcos were piloting AI, but only 57% were scaling use cases across multiple domains and only 16% characterized AI as the new normal across their organization. Only 12% reported already capturing sizable impact.
That gap should concern CEOs and CFOs. AI pilots can create activity, prototypes, and vendor excitement without creating durable operating performance. The value comes when AI is embedded into end-to-end workflows: fault detection to dispatch, customer issue to resolution, quote to provisioning, order to activation, churn signal to retention action, planning insight to capital decision.
The operating question is not how many AI use cases exist. It is which workflows AI has changed, which metrics have improved, which manual work has been removed, which exceptions are better managed, and which parts of the business can prove measurable cost, quality, speed, or customer impact.
Customer Care Is a High-Value Starting Point, but It Cannot Become a Friction Point
Customer care is one of the most visible areas for AI-enabled efficiency. Telecom operators handle high volumes of service questions, billing issues, plan changes, outages, installation questions, device problems, and retention interactions. AI can improve triage, self-service, agent support, next-best action, knowledge retrieval, call summarization, and resolution speed.
But customer care also carries brand and churn risk. Poorly designed automation can make the experience worse by trapping customers in loops, misclassifying issues, weakening empathy, or making complex problems harder to resolve. The margin benefit disappears if lower contact cost is offset by higher churn, more escalations, repeat contacts, or reputational damage.
Leaders should therefore manage customer-care AI through end-to-end outcome metrics, not simply deflection. The right measures include first-contact resolution, repeat-contact rate, escalation rate, average handle time, digital containment, save rate, complaint volume, credits issued, churn impact, and customer satisfaction by issue type. AI should reduce cost while improving the service promise, not trade one for the other.
Field Service and Truck Rolls Are Margin-Critical
Field service is one of the most important operating domains in telecom. Installations, repairs, maintenance, construction support, equipment replacement, outage response, and customer premises work all consume labor, vehicles, contractors, parts, scheduling capacity, and management attention. Every avoidable truck roll carries cost and often signals a failure upstream.
AI and analytics can improve the economics of field work through better diagnostics, remote resolution, predictive maintenance, route optimization, technician skills matching, parts planning, contractor governance, and real-time support. But those benefits require clean process design. If dispatch rules, inventory records, network data, appointment windows, and technician notes are poor, AI will automate noise rather than improve outcomes.
The strongest operators will manage field service as a closed loop. Service issues should feed root-cause analysis. Repeat truck rolls should trigger process and network investigation. Parts usage should feed replenishment and failure trends. Technician productivity should be balanced with quality and safety. Contractors should be governed through cost, quality, timeliness, and customer outcomes, not volume alone.
Autonomous Networks Need Operating Governance, Not Just Automation
Autonomous network operations are gaining momentum as operators look to reduce cost, improve resilience, and accelerate service delivery. Bain and TM Forum reported that 20% of surveyed operators had reached Level 4 or 5 maturity in select domains, while 35% expected to do so within two years. TM Forum describes autonomous networks as a path toward self-healing, self-optimizing network domains and end-to-end Zero-X operations.
The opportunity is significant. Fault management, service assurance, proactive maintenance, energy optimization, faster provisioning, private-network deployment, and real-time technician support can all benefit from more autonomous operations. But automation must be governed. Leaders need to know which decisions are automated, which require human approval, how exceptions are escalated, how performance is measured, and how risk is controlled.
Autonomous networks also require organizational change. Network, IT, service assurance, field operations, customer care, vendor management, and cybersecurity cannot operate in separate silos when closed-loop automation crosses their boundaries. The management system must define decision rights, data ownership, service standards, escalation rules, and accountability for outcomes.
Legacy Complexity Is the Hidden Cost Base
Telecom operators often carry decades of accumulated complexity: legacy copper networks, overlapping product catalogs, aging billing systems, fragmented customer platforms, duplicate processes, inconsistent service definitions, vendor sprawl, and technical debt across network and IT environments. That complexity consumes cost even when it is not visible in a single line item.
PwC notes that AI can lower service costs, reduce downtime, improve capital allocation, and raise productivity, but only when paired with product simplification, legacy shutdowns, footprint reduction, and vendor consolidation. That is an important warning. AI layered onto an overly complex operating model can make the business faster, but not necessarily simpler or more profitable.
The margin reset therefore requires simplification in parallel with automation. Leaders should identify which products, systems, sites, network assets, vendors, manual controls, and exception processes drive disproportionate cost. Then they should sequence shutdowns and migrations around customer risk, regulatory requirements, field capacity, billing impacts, and capital return.
Capital Efficiency Must Be Managed with the Same Rigor as Opex
Telecom margin improvement is often discussed through operating expense, but capital efficiency is equally important. Network buildouts, spectrum, fiber, data-center interconnect, software-defined networks, cybersecurity, and customer systems all require capital. The wrong capital allocation decisions can lock in years of depreciation without commensurate revenue.
AI can support better capital decisions by improving demand forecasting, network utilization analysis, churn prediction, capacity planning, build prioritization, and scenario modeling. But the output must feed a disciplined decision process. A model can identify opportunity; leadership still must decide where capital should be deployed, delayed, redirected, or stopped.
This requires a stronger link between finance, engineering, network operations, enterprise sales, consumer marketing, procurement, and field delivery. If these functions use different assumptions, capital governance becomes reactive. The operator needs one operating view of demand, cost-to-build, cost-to-serve, utilization, customer economics, and expected return.
The Brooks International Perspective
From Brooks International’s perspective, telecom’s margin reset is an execution-system challenge. AI can support cost control, but it will not create sustainable margin improvement unless it is embedded into the way the organization manages work, measures performance, and holds teams accountable.
The highest-value opportunities often sit at the intersections: customer care and billing, network operations and field service, product management and provisioning, capital planning and enterprise sales, vendor governance and service assurance, automation teams and front-line adoption. If those interfaces are weak, AI initiatives remain fragmented and savings fail to reach the bottom line.
Brooks International helps leadership teams translate margin strategy into daily operating performance. In telecommunications, that means building the management routines, ownership structure, data visibility, and execution cadence needed to reduce avoidable work, improve service quality, govern field activity, simplify legacy complexity, and align capital to measurable returns.
Brooks International’s view is that AI-enabled cost control should be managed as an operating capability, not a technology program. The goal is not more pilots. The goal is fewer repeat failures, fewer unnecessary truck rolls, faster provisioning, better customer resolution, smarter capital deployment, and a cost structure that can support the next phase of network demand.
What Telecommunications Leaders Should Be Asking Now
The leadership agenda should focus on whether AI and automation are improving margins through measurable operating change, not whether the company has launched more pilots.
• Which AI use cases have moved beyond pilots into scaled workflows with measurable cost, quality, speed, or customer impact?
• Can leadership trace AI-enabled savings from operational metrics to actual EBITDA, cash flow, or capital efficiency improvement?
• Where are repeat contacts, repeat truck rolls, dispatch errors, provisioning delays, billing defects, or outage patterns creating avoidable cost?
• Are autonomous network initiatives governed through clear decision rights, escalation rules, risk controls, and performance ownership?
• Which legacy products, systems, vendors, network assets, and processes should be simplified or retired to make automation valuable?
• Does the organization have one operating view of opex, capex, customer experience, service quality, and network performance?
The Leadership Imperative
Telecommunications operators cannot rely on traffic growth to solve the margin problem. The economics of the sector require a more disciplined operating model that reduces complexity, automates the right work, and links service performance to financial performance.
The winners will not be the operators with the most AI pilots. They will be the organizations that redesign work end to end, manage adoption, govern exceptions, simplify legacy environments, and prove that AI is improving both customer experience and the cost structure.
For telecommunications leaders, the mandate is clear: make AI-enabled cost control a management-system discipline, not a technology experiment.
Related Insights