Enterprises balance on-prem, cloud workloads, mobile devices and multiple access types, so edge only adds value if it reduces complexity elsewhere, rather than adds another isolated layer
In the UK, the answer sits somewhere between early promise and practical delivery. The technology foundations are far stronger than they were even three years ago, but the market is still learning where edge adds the most value, and how to deliver it at scale.
What we see is that progress over the last few years has been as much about aligning 5G networks, cloud platforms and operational models around what customers want, as it has been about advancing ultra-low latency use cases.
Realistic picture of the market
From a technology standpoint, much of the heavy lifting has already been done. Standalone 5G cores are live, networks are becoming increasingly cloud-native, and enterprises are comfortable running workloads on hyperscalers’ platforms. Progress is visible: EE’s 5G+ rollout reaches 66% of the UK’s population – more than 44 million people. The building blocks are no longer the limiting factor.
More challenging is turning edge into something that can be consistently brought to market at scale. One of the reasons adoption has progressed more gradually is that edge has often been delivered as a solution rather than a product. Deployments have involved coordinating multiple components – including mobile core, radio access, transport and cloud – partners and platforms, and the real difficulty has been stitching it all together in a way that customers can easily consume.
In parallel, customer requirements have been assessed to determine how best to meet that demand. Ultra-low latency is essential in specific scenarios, but has not been the universal driver many predicted. This is particularly true in the UK, where geography already allows centralised cloud to perform well for many use cases.
Do 5G + compute add up?
Instead, the focus has shifted towards something more fundamental: whether the 5G network and compute environment together can be trusted to do what the service requires, consistently and securely.
At the same time, many organisations are still developing the operational skills, application architectures and commercial models needed to make edge a default choice rather than a specialist deployment. AI is also beginning to influence how edge is evaluated, particularly for inference-led use cases, but for most organisations it remains an emerging consideration rather than a primary deployment driver.
What customers are asking for
When customers talk about edge, they rarely start with technology. The conversation is much more operational. They want services that work in congested environments, continue to perform under load and behave predictably in the real world – particularly where mobile connectivity is critical and conditions are variable. As one customer put it to me plainly, “It just has to work”.
Security, sovereignty and control now sit at the centre of these discussions. Organisations want to understand where their data is at any given point, how it moves across mobile and fixed networks and cloud platforms, and who is able to access or manage the systems involved. For many, especially in the public sector and regulated industries, it is not enough to say data is “in the cloud” – it needs to be built, managed and secured within the UK, end-to-end.
Edge and sovereignty
This is where edge can make a real impact, particularly for use cases that combine real-time decision-making with sensitive information. Public safety, healthcare and parts of manufacturing are early examples. These environments depend on immediate inference, reliable communications and assured 5G performance, but they also demand tight control over data and operations. Edge is relevant here as a practical way to meet those combined requirements, rather than as a theoretical architecture.
Extended reality is one example where this convergence is starting to show tangible benefits. By combining 5G connectivity with edge processing, immersive applications can become more responsive, location-flexible and usable beyond fixed sites – something that is difficult to achieve with centralised cloud alone.
Equally importantly, customers are looking for simplicity. Most organisations are already balancing on-premises systems, cloud workloads, mobile devices and multiple access types. Edge only adds value if it reduces complexity somewhere else, rather than creating another isolated layer to manage.
Making edge easier to consume
For edge to be adopted more widely and deliver the scaled, real-world benefits which customers need most, it must become easier to repeat. That means taking the complexity out of the customer experience, even if it still exists behind the scenes.
From a provider perspective, this involves doing much more of the integration upfront: aligning 5G connectivity, cloud and security components, defining clear service characteristics, and creating blueprints that can be reused across different scenarios. The aim is not to eliminate flexibility, but to avoid rebuilding the same foundations every time.
How edge fits into the network
That means rethinking how edge fits into the network itself. Low latency is a given, but the bigger emphasis should be on building trusted network environments where traffic can be monitored, audited and controlled from end to end. Mobile and fixed access should be brought together, quality of service to be engineered in, and workloads running on major platforms to be governed by the same standards customers expect from critical national infrastructure.
Cost control is just as significant. One of the challenges for edge in general has been that it can quickly become expensive and difficult to manage. To address that, providers need visibility across the entire supply chain – from network licences and spectrum through to cloud compute and application platforms – so they can work with customers to define realistic cost envelopes and then engineer the solution backwards from there.
When those core components are standardised, edge stops being a one-off project and starts to look like a service that can be adapted, scaled and reused.
How expectations may evolve
Looking ahead, enterprise expectations of networks are likely to become even more outcome-focused. Whether a service uses 5G, fibre or a hybrid of both will matter less than whether it delivers resilience, security and predictable performance.
Edge will increasingly be seen as part of a sliding scale of compute rather than a fixed destination, with workloads moving between on-premises, network edge and central cloud as requirements change. As AI-driven use cases grow, that flexibility will carry greater weight, particularly where decisions need to be made in real time.
We are also likely to see closer convergence between networks and software platforms. APIs, automation and orchestration will continue to play an important role in simplifying how services are delivered and adapted, while AI is expected to increasingly support how 5G networks themselves are optimised and operated.
For the industry, the challenge is balancing innovation with execution. Enterprises do not need every possible capability at once. They need confidence that what they deploy will integrate, scale and remain secure.
Connectivity will always be the foundation – and that means assured mobile connectivity alongside fixed access. But the real opportunity lies in turning that foundation into dependable and repeatable services that reflect how organisations operate today.
