Telcos have spent a decade and hundreds of millions of dollars chasing the 'techco' label, with little sustainable revenue growth to show for it.
Over the past several years, working with operators across EMEA and APAC, I’ve witnessed the same pattern: ambitious transformation roadmaps announced with great fanfare, followed by multi-year implementation cycles that struggle to deliver the commercial agility they promised. The industry keeps assuming the answer is a better platform, when the real problem is an unwillingness to confront what genuine business model transformation demands.
Agentic AI is the latest proposed cure. It may prove useful, but it will not rescue operators who have not first fixed the foundations beneath it.
The transformation paradox
The strategic logic was sound. Connectivity is a commodity; margin sits in digital services such as streaming, gaming, IoT, and ecommerce. Becoming a trusted aggregation hub lets telcos own the billing relationship and the data. Execution is where it breaks. Large Business Support System (BSS) programs routinely run three to five years, over budget, and often fail to deliver core objectives. In conversations with CIOs and digital strategy leaders, the pattern is familiar: a transformation program begins with the goal of enabling faster service launches, but by the second or third year, the focus shifts to stabilizing the new stack.
By go-live, the market has shifted, and the sponsoring CIO has often moved on. What remains is a fragmented architecture incapable of launching services at commercial speed.
This is the techco trap. Operators carry the complexity of a technology company without capturing its economics. Partner onboarding drags, bundling stalls, and revenue leaks between systems nobody fully owns.
Where EMEA and APAC feel it most
Living and working across these markets, the operational pressures behind techco ambition are impossible to ignore.
In the Gulf, tier-1 operators have invested heavily in digital lifestyle portfolios, but BSS layers were never built for multi-currency billing, real-time partner revenue sharing, or multi-jurisdiction tax compliance. The services exist. The monetization engine does not.
In Southeast Asia, hyper-competition and multi-SIM behavior compress the window for capturing a subscriber's digital wallet to months. Operators in Indonesia or the Philippines cannot wait three years to launch a bundled offer. Speed to market decides the outcome.
The constraint the industry must hear clearly: agentic AI is only as good as the data it operates on.
Revenue velocity is the metric that matters
The industry must retire the multi-year program as the unit of strategic progress. The metric that matters is revenue velocity: how quickly a new service moves from decision to live offer. For most telcos, the cycle runs for many months. Operators with modular, API-driven monetization infrastructure do it in weeks.
Bundling shows the gap clearly. Parks Associates data confirms streaming video is now the most common broadband add-on globally. The operators capturing that opportunity are not those with the most sophisticated BSS, but those who can activate a partnership, configure pricing, and place a personalized offer in front of the right segment within weeks. The same applies across IoT, gaming, and insurance: run experiments quickly, iterate pricing, kill what fails.
Agentic AI: Useful tool, not a foundation
Point automation inside individual systems delivered efficiency but not integration. You end up with faster silos. Agentic AI is different because it orchestrates actions across systems. A retention agent can pull churn signals from billing, check entitlements, select a personalized bundle, execute the offer, and log the outcome autonomously, without a human connecting four platforms.
The constraint the industry must hear clearly: agentic AI is only as good as the data it operates on. Deploy agents on fragmented, poorly governed data, and you won’t get slower wrong answers. You’ll get fast, confidently wrong ones at scale. The prerequisite is clean subscriber data, clear system boundaries, well-defined entitlement logic, and accurate revenue attribution. That is not a reason to delay AI investment. It is a reason to sequence it correctly.
I see this directly. The operators moving fastest on agentic capabilities are not those with the loudest AI ambitions. They are the ones who did the unglamorous work of cleaning subscriber data and consolidating entitlement logic first.
Stop pricing your way out of a trust problem
When transformation fails to generate new revenue, the default response is to raise prices on existing services. It is simple and shows up next quarter. It is also the fastest way to destroy the trust that a digital aggregation strategy depends on. An operator that cannot offer flexible digital services but can reliably raise broadband prices is not a techco. It is a utility borrowing a marketing vocabulary, which it cannot operationalize. The subscribers most likely to churn are the high-ARPU, digitally engaged segments that every growth strategy targets.
The operators that will define the next decade are building something different: the ability to deliver relevant, fairly priced offers that each subscriber can act on easily. That requires clean data, flexible monetization infrastructure, and the will to treat subscriber intelligence as a core competency.
Any operator running a BSS transformation for more than two years that still cannot launch a new digital service in under eight weeks is already out of the race. Agentic AI can accelerate operators who have built the right foundations. It cannot substitute for them.
