If CapEx is not tied to outcomes, you are spending blind

Connecting network investment to business outcomes

Network investment is one of the most critical decisions telecom operators make. Yet planning across RAN, FTTH/FWA, Transport, Core and Data Centers is still often fragmented, with engineering, commercial and finance teams working in separate tools and measuring success against different KPIs.

This slows planning cycles, weakens visibility of ROI and makes it harder to answer the questions that matter most to the business: where to invest first, what payback to expect, how customer experience will improve and where new revenue or market share could be unlocked.

As capital requirements increase, operators need to move from technical network planning to business-driven capital allocation. That means aligning CTO, CFO and CMO priorities, so investment decisions reflect customer impact, financial returns and growth opportunity, not only network performance.

A multi-agent Smart CapEx platform for end-to-end planning

The Catalyst project, E2E multi-agent smart network CapEx - Phase III, introduces an end-to-end multi-agent AI Smart CapEx platform that connects technical planning with financial outcomes.

AI agents analyze demand forecasting, coverage gaps, capacity constraints, churn risk, competitive white spaces and ROI across RAN, FTTH/FWA, Transport, Core and Data Centers. These inputs are combined to produce explainable investment priorities based on business value, rather than technical KPIs alone.

The project extends previous Smart CapEx phases by integrating three key TM Forum transformation pillars:

• Autonomous Network Levels (ANL) as a Service for maturity assessment and investment prioritization cross-domain
• Data as a Product (DaaP) for AI-ready, governed and reusable data products
• Multi-Agent Intent-Based Orchestration for autonomous execution from planning to deployment

This enables operators to move from disconnected planning toward a unified model where technical, commercial, and financial priorities are aligned around measurable outcomes.

Improving CapEx efficiency, customer experience and rollout decisions

The expected impact is both strategic and operational. The project team estimates that AI-driven Smart CapEx could improve CapEx efficiency by 15–20% and accelerate planning cycles by 30–50%, while supporting better EBITDA, faster payback and stronger alignment between network, finance and marketing teams.

For operators, this means investment decisions can be prioritized around the highest combination of ROI, customer experience improvement, churn reduction, and growth potential.

For the industry, the Catalyst provides a practical model for linking network investment directly to business performance using standardized APIs, intent models and agent-based automation.

For society, more efficient allocation of resources can support faster rollout in underserved areas, improve digital inclusion, and enable more sustainable infrastructure investment.

Guilherme Fernandes Silveira, Engineering Director at Telefonica Vivo, says the Catalyst is valuable because it “helps us connect network investment directly to customer experience and business value.” For Telefonica Vivo, he adds, improving where and how CapEx is invested is critical: “AI-driven decision-making helps ensure every investment delivers stronger customer outcomes, better returns and faster connectivity expansion across our markets.”

DTW Ignite 2026

At DTW Ignite 2026, E2E multi-agent smart network CapEx – Phase III demonstrated how AI can evaluate network investment scenarios across multiple domains and generate prioritized recommendations based on ROI, customer experience and growth potential. The showcase highlighted how decision-making can shift from siloed planning to a unified, data-driven approach. The project was named winner of the Open Innovation Catalyst award for TM Forum asset excellence.