Summary

  • e& has announced a target to grow international connectivity capacity from a stated 20 Tbps to more than 500 Tbps by 2030, but its public announcement does not define a reproducible measurement perimeter.
  • Until the company identifies the baseline date, capacity stage, geography, ownership basis and counting method, readers cannot tell whether later figures represent comparable, customer-usable capacity or simply a larger headline.

The first question about e&’s 500 Tbps ambition is not how large the number sounds. It is what a future auditor would have to count to reproduce it. In an announcement dated 5 October 2026, the company said it planned to increase international connectivity capacity to more than 500 Tbps by 2030, describing that as more than 25 times current capacity. A quoted statement put the starting point at 20 Tbps. Those figures establish the target e& has announced. The release does not provide the dated baseline, technical perimeter or counting rules needed to test progress against it.

That omission matters because “capacity” can describe several different states. A system may have a designed capacity, equipment installed to support some of it, capacity lit on the network, capacity provisioned for a service, or capacity that a customer can actually order under defined conditions. These are not interchangeable. A project can add theoretical or installed capacity before it becomes usable; conversely, service capacity can depend on arrangements that are not visible from a physical cable map. e& has not stated in the reviewed announcement which stage its 20 Tbps baseline or 500 Tbps target represents.

Assigning either figure to a stage would therefore go beyond the evidence.

The geography and measurement points are equally consequential. “International” might be bounded by the UAE border, by named landing stations, by particular network endpoints or by a wider set of markets. The result could also differ depending on whether a figure is measured at the cable, at a gateway, across a backbone, or at a customer-facing service boundary. A reproducible metric needs a fixed perimeter: what locations and links qualify, where throughput is measured, and how the operator aggregates those values. Without that perimeter, a later update could use the same label while counting a different network footprint.

The company’s own Capacity Hub page illustrates why a route list cannot substitute for a KPI definition. It describes access to more than 20 terrestrial and submarine cable systems and wholesale services on systems that do not necessarily land at e&’s own cable stations. That is useful context about the commercial footprint, but the page does not say that every listed system contributes to the 20 Tbps baseline or the 500 Tbps target. Nor does it disclose the capacity amount assigned to each system or the point at which it is counted. It would be incorrect to infer the target’s composition from the product page.

Rights to use infrastructure raise another accounting question. e&’s 2025 integrated annual report describes an indefeasible right of use, or IRU, as a fixed-term contractual right to use capacity on a transmission cable. That explains a category of access, not the treatment of IRUs in this target. An auditable disclosure should distinguish capacity owned directly from capacity available through an IRU, lease, consortium or other partner arrangement. Those forms can all support service delivery, but their duration, control and renewal exposure differ.

Counting them together may be reasonable for a measure of serviceable capacity, provided the basis is disclosed and unchanged. Treating them as equivalent without explanation would obscure what e& controls and what it can continue to access.

Directionality and duplicate counting also need explicit rules. A link’s capacity could be described per direction or as a combined bidirectional figure. A shared system might appear in more than one market or endpoint calculation. Capacity purchased through a partner might also be represented in a system-level figure. None of these possibilities proves that e& has double-counted anything; they show why the aggregation method should be published. The same methods must be applied to the baseline and every subsequent checkpoint for the ratio between 20 and 500 Tbps to mean anything over time.

Regulation does not fill this corporate reporting gap. The UAE Telecommunications and Digital Government Regulatory Authority’s 2022 international cable rules distinguish cables that transit, land, connect or terminate and set out a licensing and permitting framework. They clarify regulatory concepts around cable works. The reviewed rules do not prescribe a method for measuring e&’s 500 Tbps company target, and they are not evidence that any particular target component has been permitted or delivered.

A disclosed regulatory filing could corroborate a specific milestone if it covered the same asset and stage, but the rules themselves do not validate the headline.

Cloudflare’s use of the same 500 Tbps magnitude offers a narrow methodological comparison. Cloudflare says its figure represents provisioned external interconnection capacity and distinguishes it from peak traffic. That is a useful example of a company stating what its number measures; it is not a definition that can be transferred to e&. The networks, service boundaries and purposes differ. Capacity is also not traffic: a network’s ability to carry data is not the amount customers are using at a given moment.

For e&’s target to become a credible public commitment, the company would need to publish a dated 20 Tbps baseline, hold the perimeter constant, identify the measurement layer and define which capacity stages count. It should state how owned, IRU, leased and partner capacity are treated; whether figures are per direction or combined; how shared capacity is de-duplicated; and which customer-available evidence supports each annual update. Independent assurance would not make the target achievable by itself, but it could make reported progress reproducible.

Until those rules are public, the 2030 number remains a company ambition rather than a verifiable delivery record.

Sources