News · Technology

Networking Revenue Growth Leaves Business Returns Unproven

By AWEI · AI-compiled · Published 2026-09-16 · 2 sources · forbesjapan.com, www.iccsz.com

Networking revenue reflects equipment demand; marketing surveys measure responses. Neither establishes durable business returns.

For teams committing technology budgets, two striking percentages answer different questions: 94% describes reported growth in supplier revenue from direct cloud-provider sales; 68% describes a cited share of business buyers already having a vendor in mind. ICC reports the first, while a Forbes JAPAN contribution cites the second. Neither establishes that an organization buying AI capabilities will earn a durable return. Their connection is the investment question running through the technology cycle: who records revenue now, and who still has to demonstrate value?

A percentage of what?

ICC’s September 11 account of Dell’Oro Group research reports 25% year-on-year growth in high-end routing and aggregation equipment revenue during Q2 2026. Direct cloud-provider sales grew 94%, edge routers 20% and enterprise high-end routers 31%. Those are revenue comparisons across different segments, not interchangeable measures of adoption. Without absolute revenues, the largest growth percentage cannot reveal the largest contribution to total expansion. Revenue also combines purchasing quantities and what buyers pay, leaving prices and product mix as possible contributors alongside increased equipment demand.

ICC describes double-digit expansion across customer groups and major regions, with North America and Latin America growing fastest. That breadth makes the account relevant beyond a single customer category, but it does not isolate AI as the cause of every purchase. Purchasing cycles could explain part of the increase, and direct-cloud growth could include changes in sales channels. The important distinction is between suppliers recording business and customers realizing useful capacity. A sale establishes a transaction; utilization, implementation costs and subsequent benefits require different evidence.

Better data is a strategic hypothesis

In Forbes JAPAN’s September 12 contribution, Rob Sanchez recommends refreshing first-party data, scrutinizing outside providers, sharing information between sales and marketing, and auditing practices at least annually. His proposed competitive mechanism is a feedback loop: more relevant information could improve decisions, whose results could inform subsequent improvements. Distinctive data could therefore matter beyond access to widely available AI tools. But this remains a strategic hypothesis. The article does not demonstrate that these practices created a durable advantage, or separate their effects from management quality and existing customer relationships.

The survey percentages require the same denominator discipline as equipment sales. Forbes JAPAN cites Anteriad findings that 57% of agile marketers and 40% of marketers overall expressed confidence in using appropriate data. Confidence is a reported assessment, not measured profitability. Its Forrester comparison places 19% of marketing leaders who think buyers begin with a clear vendor preference alongside 68% of buyers reporting an existing preference. Different respondent groups and constructs may expose an assumption worth investigating, but they are not before-and-after results. The cited Gartner figure of 65% technology-enabled, data-driven B2B sales organizations remains a forecast, even with 2026 as its target year.

Capacity and value arrive at different stages

The WTO’s Global Trade Outlook and Statistics, October 2025, provides a value-chain lens for separating infrastructure investment from productive use. Its Secretariat estimates concern global merchandise trade across AI-related product lines in the first half of 2025, with a 2025–2026 outlook. They are trade analysis and projections, not measurements of Q2 2026 router returns or marketing performance. Overlapping tariff preparation and other economic conditions limit causal attribution. Applied conceptually, the report helps explain why spending on networking and cloud capacity can precede evidence of customer value; it does not corroborate these particular revenue figures.

That timing creates an asymmetry. Suppliers may receive sales revenue while purchasers bear implementation costs and uncertain returns. Marketing teams could improve results through better coordination, but established vendor relationships could also explain buyer preferences without an AI data advantage. Similarly, capable organizations may both report greater confidence and perform better, making confidence a companion to performance rather than its cause. The wider business issue is how value is distributed along an investment chain. These two articles provide no basis for concluding that every participant benefits, or that equipment expansion proves economy-wide productivity growth.

The watchpoint is sustained equipment demand accompanied by documented utilization and net customer benefits. That combination would strengthen the infrastructure-to-value interpretation; fading growth without such benefits would give a purchasing-cycle explanation more weight. The marketing thesis needs comparable outcomes around changes in data practices, with other interventions accounted for. Until those results appear, the repeated question remains useful: a percentage of what? Supplier revenue, respondent confidence, buyer preferences and forecasts each describe something real within their stated bounds, but this leaves lasting business returns unproven.

All newsNews archive