Sercomm's record second-quarter 2026 revenue didn't come from selling more broadband routers - it came from selling far less commoditised 5G, edge AI and IoT hardware instead. For UK businesses, that pivot is a preview of where connectivity spending is heading over the next 18 months: away from bandwidth alone and toward smart, on-site processing power. In short, UK organisations that still treat 5G as just "faster broadband" risk overpaying for infrastructure that under-delivers, while those adopting edge AI-enabled hardware can cut cloud costs and unlock real-time automation this year.
What is the Concept
Sercomm is a Taiwan-based original design manufacturer (ODM) that has spent two decades building the broadband routers and gateways used inside internet service providers' networks worldwide, including hardware historically deployed by UK providers such as BT, Sky and TalkTalk. Its 2Q26 results show revenue increasingly driven by 5G fixed wireless access (FWA) units, edge AI gateways and IoT modules rather than traditional fibre and DSL routers.
Edge AI means processing data on the device or at a local network node instead of routing everything to a distant cloud data centre. Paired with 5G's low latency and IoT sensors collecting real-time data, this forms what we call the Edge Ladder Model: Connect (5G or fibre), Compute (edge AI processing on-site), and Control (IoT devices acting on that processing automatically). Sercomm's shift shows hardware vendors now building for the top two rungs of that ladder, not just the first.
Why It Matters in United Kingdom (2025–2026 Context)
UK 5G coverage from Vodafone, EE, Three and Virgin Media O2 has matured well beyond London and Manchester into secondary cities such as Bristol, Leeds and Birmingham, but adoption by SMEs remains patchy because most firms still treat it as a mobile-only upgrade. Ofcom's ongoing spectrum releases and the push toward full fibre switch-off by 2027 mean UK businesses will soon need a connectivity strategy that goes beyond simply swapping one broadband line for another.
Rising energy and cloud hosting costs have made every pound of infrastructure spend matter more in 2026. A hardware vendor the size of Sercomm reorienting its entire product line toward edge AI and IoT is a strong market signal: UK operators will increasingly ship CPE (customer premises equipment) that does local computation by default, and businesses that don't plan for this will pay for capability they never use.
How AI Is Changing This
The contrarian insight here is that the value of 5G for most UK businesses was never really the speed - it was the latency low enough to make edge AI viable. A warehouse in Sheffield or a retail chain in Leeds doesn't need faster downloads; it needs a camera or sensor that can make a decision in milliseconds without waiting on a round trip to a cloud server. Edge AI gateways, the exact category Sercomm is now scaling, make that possible on ordinary business premises.
This also changes the economics of automation. Instead of paying recurring cloud compute fees to process every video frame or sensor reading, UK firms can push inference to the edge device itself, paying once for the hardware rather than continuously for cloud cycles. For SMEs watching every line of their operating budget, that shift from opex to a fixed, one-off capital cost is often the more attractive path.
Real-World Examples
Consider a mid-sized logistics operator running a distribution centre near Birmingham. Historically it would lease a fixed line for connectivity and send warehouse camera feeds to a cloud AI service to flag damaged pallets, paying both a broadband bill and a per-frame processing fee. Under the Edge Ladder Model now being built into hardware like Sercomm's, the same site could run 5G FWA as its primary connection and process the camera feed locally on an edge AI gateway, only sending exceptions to the cloud.
UK retailers are following a similar pattern with IoT-enabled stock sensors in stores across London and Manchester, where local processing flags shelf gaps in real time instead of waiting for a nightly cloud batch job. These aren't hypothetical use cases; they are the exact product categories vendors like Sercomm reported record demand for in 2Q26.
Practical Insights / Actions
The most common founder mistake in the UK right now is signing a 5G or fibre upgrade contract and assuming the connectivity itself will solve automation and speed problems. It won't. The hidden opportunity is on the compute layer: pairing that connection with an edge AI gateway is what actually reduces recurring cloud bills and speeds up decision-making on-site. Before upgrading connectivity alone, UK founders and CTOs should map out where their business generates real-time data - stock, footfall, machinery, vehicles - because that's where edge AI delivers the fastest return.
This is exactly where a technology partner adds value. RP SoftTech works with UK SMEs to design and integrate edge AI and IoT systems on top of existing 5G or fibre connections, so businesses avoid buying capability they can't use and instead build a connectivity stack that pays for itself through reduced cloud spend and automation gains.
Future Outlook
Expect more hardware vendors serving UK operators to follow Sercomm's lead through 2026 and 2027, shipping CPE with built-in AI inference as standard rather than optional. For UK businesses, that means the connectivity decision made this year - fibre versus 5G FWA, cloud-only versus edge-enabled - will shape operating costs and automation capability for several years, not just months.
Businesses that get ahead of this shift now, while edge AI hardware is still a differentiator rather than a baseline expectation, stand to build a meaningful cost and speed advantage over slower-moving competitors in their sector.
Conclusion
Sercomm's record 2Q26 revenue is a hardware story on the surface, but for UK businesses it's really a signal about where connectivity spending and cloud cost savings are heading in 2026. Firms that pair 5G with edge AI and IoT, rather than treating connectivity as a standalone upgrade, will be the ones capturing the cost and speed advantages first. If you're planning a connectivity or automation upgrade this year, an audit of where edge AI fits into your existing infrastructure is the logical next step - and it's a conversation RP SoftTech is set up to have.