This article first appeared in UC Advanced magazine issue#21.
Kim Holder, director – enterprise unified communications, Colt Technology Services discusses how UK providers can deliver consistent quality for voice, video and AI-enabled customer contact everywhere
Every few years the way we work shifts, and the strain shows up first in real-time media. Cloud calling is replacing PBXs, video has become the default meeting format, and collaboration and contact centres are now adding real-time transcription, meeting summaries, diagnostics and agent assist. These features are no longer optional, they are part of daily business. The question for UK providers is whether the network can deliver consistent quality for Teams, Webex, Zoom, Genesys Cloud, NICE and others, not only in London but also across regional sites and home offices.
The UK today
There is plenty to respect in the UK’s position. Gigabit access has expanded quickly, London remains one of the most interconnected hubs in the world, and the data centre and cloud ecosystem is deep. Many enterprises already run large estates on Teams or Webex, and cloud contact centres are ubiquitous.
Where users still feel pain is consistency of experience. Outside the biggest metros, options for resilient, enterprise-grade fibre and diverse paths can be limited or costly. At busy times, latency and jitter vary. Users do not file tickets that say, “jitter increased by 12 ms”, they say “the call sounded poor” or “my meeting froze”. That variability is the gap to close if the UK is to offer the same experience everywhere that it already delivers in the capital.
What good looks like
For collaboration and contact, raw bandwidth is rarely the constraint. The real issues are path, peering and predictability.
✅ Path. Media should take the shortest, cleanest route to the vendor’s nearest media edge. If traffic hairpins through the wrong region, you pay in delay, echo and talk-over.
✅ Peering. Direct, well-engineered interconnects between your carrier and the platforms you use are critical. Poor peering shows up as random drops that no amount of extra bandwidth fixes.
✅ Predictability. Low latency, jitter and loss matter more than raw speed. It’s essential to keeping calls stable, video smooth, and transcription clean.
A solid design is straightforward. Use SBCs to control media and apply policy. Deploy SIP from the cloud with consistent numbering, emergency calling and compliance recording. Route real-time traffic through SD-WAN or SASE to the nearest media node, with failover that does not drop calls.
These are the basics that underpin every successful deployment. They set the stage for the next layer of change, where AI-driven features are starting to reshape how collaboration and contact centres behave and what the network must carry.
AI and traffic load
AI collaboration benefits are already becoming mainstream. Features already in use include live captions and translation, acoustic echo control, meeting and call summaries, quality diagnostics that spot a bad local network, agent assist that surfaces the right policy while the customer is speaking. These functions clearly improve outcomes for users and customers, but they also change the traffic profile. Some processing can happen on the device, some relies on media nodes close to the user, and some runs best in cloud regions. The result is more paths to manage, and more demand for stable, low-latency connectivity to keep experiences consistent.
What is changing in the UK?
Investment continues across the ecosystem. Carriers are extending fibre into more locations, lifting backbone capacity and improving diverse routing. Vendors are placing more media nodes closer to users, which helps a great deal. Government programmes for hard-to-reach areas are making progress, although it takes time.
Colt is expanding its European fibre footprint, deepening interconnects to major collaboration and CCaaS platforms, and operating SIP trunks and SBCs as managed services with defined media quality targets. Upgrades on international routes, including trials of very high-capacity services, show how far performance can scale. The goal is not just more speed but consistent quality for voice and video, with routing that avoids unnecessary detours and resilience that maintains service during faults.
What can be done now?
You do not need a grand transformation to lift quality. You need to do the basics well.
1 Measure what users feel. Track call setup success, post-dial delay, MOS, jitter and loss per site. Use synthetic probes from branches and home offices. Fix route and peering issues, not just bandwidth.
2 Rationalise the underlay. Reduce provider sprawl and concentrate traffic with carriers that can evidence strong peering to Microsoft, Cisco, Zoom and your CCaaS of choice. Ask for route maps and evidence.
3 SBC placement. Keep media local where it helps performance, centralise when policy requires it, and test failover under live conditions. Resilience should support both compliance and user experience.
4 Treat numbering and compliance as part of quality. Porting delays, brittle recording or poor survivability undermine performance gains. Handle numbering, emergency calling and recording with the same priority as latency.
5 Design for hybrid work and change. Assume a mixed estate of offices, home workers and acquired sites. Make policy and media quality follow the user so locations feel the same.
A balanced outlook
The UK is not starting from scratch, nor is it finished. The core is strong, the partner ecosystem is deep, and the direction of travel is right. The task is to smooth the remaining rough edges so that a branch in the Midlands, a contact centre in the North and a hub in London all have the same experience on a given afternoon.
Colt is playing its part by providing the fibre and routing between cities, strong interconnects to collaboration and CCaaS platforms, and managed voice services that behave well under pressure. The task for providers is to turn that capability into a dependable experience for employees and customers.
Do we have what it takes?
Yes, but only if we keep the focus on the basics that matter most to users. Short, clean paths, strong peering, predictable performance, and clear, reliable delivery of voice, video and AI features. If providers stay disciplined on those points, the UK will be well placed to deliver consistent quality across all locations.
