The point of view
The scarce asset keeps moving. The enterprise must be able to move with it.
Technology, media and telecommunications companies increasingly compete across the same stack: compute, intelligence, software, content, distribution, connectivity and trust. Their economics remain distinct. A software workflow, a sports right and a fiber route cannot be valued with the same growth story.
Koko’s hypothesis: by 2030, the strongest TMT enterprises will manage a portfolio of capabilities, rights and relationships. By 2035, their advantage could depend as much on how quickly they can reconfigure that portfolio as on the assets they currently own. A capability is a repeatable ability to deliver an outcome, with people, knowledge, technology, authority and economics attached.
That is a management choice to test—not an inevitable industry destination. This sector reading extends the Enterprise Future Guide; its companion, TMT CFO Future Guide, asks how to protect, create and orchestrate the resulting value. Evidence reviewed September 16, 2026. Forecasts remain forecasts; transaction status is dated.
Why consolidation and separation can both make sense
Recent research points toward control of scarce capabilities, while also arguing for sharper portfolio boundaries. KPMG emphasizes strategic buyers and monetization. PwC considers ways to own or access capability. McKinsey explicitly treats media consolidation, separation and partnerships as simultaneous responses to structural change. KPMG ↗ PwC ↗ McKinsey ↗
Koko’s inference is that convergence of capabilities does not require convergence of ownership. A company may need shared identity, content rights and distribution without owning every partner. It may need dependable compute without owning a data center. Conversely, access alone may be inadequate where a partner controls the customer, pricing or critical knowledge.
Start with the constraint on the customer outcome. Then choose whether to build, buy, partner, license, reserve capacity or divest. Revisit the choice when the constraint moves. Acquisition is one implementation of strategy; it should not become the strategy itself.
Three sectors. Three different tests of value.
| Sector | What is changing | Koko’s 2030 hypothesis | Evidence that would weaken it |
|---|---|---|---|
| Technology | Agents may change how software is accessed and priced; inference and reliability become operating costs. | Durable advantage concentrates in proprietary context, embedded workflows, distribution and verifiable outcomes. The interface becomes easier to replace than the underlying capability. | Customers retain seat-based buying, context proves easy to migrate, or agent support costs overwhelm willingness to pay. |
| Media and entertainment | More content competes for limited attention; creators, streaming, gaming and live experiences share audiences and IP. | Rights, trusted brands and the ability to turn attention into recurring customer value matter more than the volume of content produced. | Cheap undifferentiated content sustains retention and margins, or exclusive rights fail to earn their acquisition and production costs. |
| Telecommunications | AI workloads, fiber convergence, network APIs, edge and satellite partnerships expand potential uses of connectivity. | Selected operators coordinate dependable business outcomes across networks and partners; others earn stronger returns by specializing in infrastructure. | Buyers resist outcome pricing, API demand stays small, or partner costs consume the incremental margin. |
The evidence supports different starting conditions: Gartner’s software repricing forecast, EY’s media portfolio and authenticity arguments, and PwC’s telecom emphasis on footprint economics. It does not prove Koko’s hypotheses. Gartner ↗ EY ↗ PwC ↗
Technology: the software business may outlive the software interface
Headless SaaS separates a system’s capabilities from the screens people use. An agentic orchestration layer can coordinate work across those capabilities. Neither concept automatically removes the need for a system of record, commercial licenses or accountable owners.
Gartner forecasts pressure on application spending as agents change usage. The useful counterargument is a16z’s view that incumbents can benefit from their embedded data and workflows. Both can be true: an incumbent may remain essential while its pricing unit changes. Gartner ↗ Andreessen Horowitz ↗
Koko’s test is specific: who owns the context, permissions and accepted outcome when an agent crosses five products? Build a knowledge product that records definitions, provenance, contractual rights, permissible actions and exceptions. Make it portable enough to change the orchestration layer. Let the CTO/CIO manage architecture, the CDO govern meaning, the CAIO evaluate agents, the CISO control identities and the CFO verify economics.
Semiconductors, advanced packaging, sovereign compute and physical AI introduce physical constraints. Deloitte’s predictions and Bessemer’s infrastructure thesis point toward orchestration and dependable infrastructure as enabling work. They do not establish that every new workload will pay for its capacity. Deloitte ↗ Bessemer Venture Partners ↗
Act now: select one valuable workflow. Measure contribution per accepted outcome, including inference, licenses, human review, remediation and customer support. Compare the current application, a headless configuration and an agent-mediated alternative before committing to a platform-wide redesign.
Media: a rights portfolio needs an operating system
A media company can automate production and still lose the customer relationship. It can own valuable IP and lack the rights to reuse a likeness, soundtrack or archive in a new geography, format or AI workflow.
EY emphasizes simplification, experiences and creator collaboration. PwC provides a broader market outlook. Koko’s next step is to connect rights, audience evidence and unit economics at the level where decisions occur. EY ↗ PwC ↗
Imagine a franchise capability team spanning creative leadership, product, commercial, legal, data and finance. Its knowledge product records chain of title, territories, windows, consent, partner obligations and permitted uses. Agents can propose localization, marketing variants or licensing opportunities inside those boundaries. Named people retain editorial judgment and authority for material rights decisions.
Creator partnerships and sports rights are different commitments. Assess minimum guarantees, production obligations, rev shares, renewal leverage and access to audience data. More views do not establish incremental contribution; a bundle can improve retention while subsidizing an uneconomic right.
Act now: choose one franchise or content library. Connect its rights register to distribution and revenue data. Test a new format or partner arrangement against incremental cash contribution and rights exceptions, with a human approval gate for new uses.
Telecom: expose the network, keep responsibility explicit
Accenture and Capgemini describe enterprise demand for more than connectivity. Analysys Mason’s public prediction summary highlights the pressure of AI infrastructure capital requirements. The opportunity and the funding problem belong in the same conversation. Accenture ↗ Capgemini ↗ Analysys Mason ↗
Network APIs can expose capabilities such as identity verification to outside developers. Aduna’s July launch is a concrete product signal, not proof of adoption or profits. The carriers’ May satellite announcement is an agreement in principle, not a completed joint venture or universal coverage. Aduna ↗ AT&T ↗
TM Forum’s architecture direction and GSMA’s mobile AI vision suggest ways to connect network, edge, device and cloud. Koko’s condition is accountable service delivery across that chain. Autonomous operations need bounded permissions, escalation, recoverability and evidence of customer impact. TM Forum ↗ GSMA ↗
Act now: test one network capability with a paying design partner. Make one team answerable for service quality, security, partner handoffs, billing and contribution. Compare build, wholesale access and partnership options. Treat 6G and quantum-enabled services as staged options until performance and demand justify scaling; connect to the core guide’s frontier scenarios.
The executive partnership becomes a capability charter
Organizing around capabilities does not eliminate functional expertise, legal entities or professional accountability. It changes how work and funding cross their boundaries.
| Capability | Executive partnership | Shared decision and evidence |
|---|---|---|
| Trusted monetization | Product and commercial leaders + CFO + CDO + general counsel | Define the paid outcome, allowable data/rights use, attribution and net contribution. |
| Dependable intelligence | CTO/CIO + CAIO + CISO + CDO | Set architecture, knowledge quality, agent permissions, evaluations, escalation and recovery. |
| Flexible portfolio | CEO + CFO + strategy/M&A + procurement + legal | Choose ownership or access; record exclusivity, minimum commitments, control and exit rights. |
| Workforce adaptation | CHRO + capability owner + finance + professional leaders | Redesign work, assess proficiency and redeploy capacity; verify service quality and benefits. |
| Assurance | Accountable executives + risk/compliance, with independent internal audit | Demonstrate controls and challenge claims without making the delivery team its own independent assessor. |
Here, chief product and chief procurement officers have separate mandates even where both use the abbreviation CPO. Converging responsibilities demand clearer decision rights. A charter should name the outcome owner, budget, delegated actions, evidence threshold and escalation route.
Where the published views align—and where Koko goes further
This is a comparison of reviewed arguments, not a vote of independent studies. Multiple publications can share data; company announcements have different evidentiary weight from research.
| Perspective | Distinctive emphasis | Koko’s extension or challenge |
|---|---|---|
| KPMG / PwC | Strategic capability access and monetization. KPMG ↗ PwC ↗ | Identify the scarce constraint before choosing the transaction structure. |
| McKinsey | Consolidation, separation and partnership can coexist. McKinsey ↗ | Design capability boundaries so the enterprise can both integrate and separate. |
| Bain / BCG | Integration plus transformation strains execution; deal recovery can be concentrated. Bain & Company ↗ BCG ↗ | Put management capacity and downside liquidity beside the synergy case. |
| Deloitte | Interdependent technology advances need orchestration and enabling infrastructure. Deloitte ↗ | Do not fund every layer equally; locate the bottleneck for each customer outcome. |
| EY | Media trust and portfolio choices; telecom transformation and security risks. EY ↗ EY ↗ | Make rights and trusted execution operating capabilities with measurable economics. |
| Accenture / Capgemini | Telecom B2B outcomes and ecosystem delivery. Accenture ↗ Capgemini ↗ | Stated customer expectations must become paid, profitable delivery. |
| Gartner / a16z | Application pricing exposure versus incumbent context and workflow advantage. Gartner ↗ Andreessen Horowitz ↗ | Test repricing and incumbent adaptation together; they are not mutually exclusive. |
| Bessemer / TM Forum / GSMA | Infrastructure investment thesis and industry architecture directions. Bessemer Venture Partners ↗ TM Forum ↗ GSMA ↗ | An investable thesis or shared architecture is not evidence of customer demand. |
The report selector below lets you choose two to six publications, inspect their access limits and ask Koko to compare them. The wider libraries retain the other Future Guide research.
Enterprise 2035: four conditions, different choices
These scenarios extend the parent guide’s two axes: commercially useful technology and enterprise readiness. They have no assigned probabilities.
| Scenario | TMT expression | What changes the decision |
|---|---|---|
| Compounding enterprise | Useful agents and strong foundations connect software outcomes, licensed media experiences and dependable network services. | Scale capabilities with verified contribution; buy scarce assets only when ownership outperforms access. |
| Stranded intelligence | Technology works, but incompatible rights, definitions and incentives block cross-company coordination. | Pause duplicate agents and acquisitions that amplify complexity; repair knowledge, contracts and operating authority. |
| Disciplined optionality | Foundations are strong but demand, reliability or economics develop more slowly. | Improve current products, license selectively and preserve cash; keep reversible learning options. |
| Fragmented progress | Weak readiness combines with constrained technology, supply or regional interoperability. | Secure essential services and local compliance; simplify the portfolio and limit irreversible capacity commitments. |
A sovereignty shock can affect any scenario. The EU Digital Networks Act remains a proposal in the reviewed evidence, illustrating why policy direction and enacted obligations must be tracked separately. European Commission ↗
Start with one boundary, one capability and one decision
In the first 30 days, map the dependencies behind one material customer outcome: data, IP, people, platforms, compute, connectivity and counterparties. In days 31–60, assign the capability charter and test an alternative ownership or partnership arrangement. In days 61–90, decide whether to scale, renegotiate, separate or stop using agreed economic and control evidence.
The falsifiable question is whether this design produces better outcomes with less coordination cost. If it merely adds another committee, change it.
For transaction examples, status and the capital playbook, continue to TMT CFO Future Guide. For the broader model, return to C-suite partnerships. This is a dated research edition; material deal, regulatory, pricing and operating evidence should trigger an editorial review.