David Trossell, CEO & CTO of Bridgeworks speaks to Digitalisation World about “Distance Tax” and the growing performance degradation on wide area network security strategies.
September 7, 2026
Latency and packet loss, not bandwidth, are the hidden “Distance Tax” undermining enterprise networks — and experts argue AI-driven WAN acceleration, not costly upgrades, is the smarter fix.
Enterprises keep buying faster connections, but distance, latency and packet loss can leave expensive bandwidth underused. The smarter question is not how wide the pipe is, but how much data can pass through it – particularly in this digitally transforming era. That’s because network capacity is only valuable when organisations can use it.
Over recent years there has been much coverage about organisations moving towards digital transformation: the challenges they face and the opportunities it presents. However, there is one crucial factor that could stop some of these worthy projects in their tracks – the Distance Tax. This is commonly caused by Wide Area Network (WAN) latency and packet loss, and so they can subsequently become a business resilience issue – affecting infrastructure returns on investment.
Tim Glatz, Head of Interconnection at DataBank, writes in his article for DataBank Digest, ‘The Hidden Costs of Network Latency (and How Interconnections Can Help):
“Network latency has quietly shifted from an operational metric to a strategic risk. For years, IT teams measured performance almost entirely in uptime, watching for outages, tracking service level agreements and treating five nines on a dashboard as proof the network was doing its job. That measure is no longer sufficient. A network can be fully available and still be costing the business money every day, one slow response at a time.”
Increasing with distance
This ‘tax’ increases with distance. That’s why organisations are often looking to the Edge, and it is also why many of their disaster recovery sites are placed precariously within the same circles of disruption (which could put their data and operations at risk during a natural or manmade incident). Instead, they should still look to placing their data centres and disasters recovery sites in disparate locations situated far away from each other. Latency and packet loss needn’t stop them from doing this.
Another response to latency and packet loss is for them to invest heavily in faster fibre, cloud platforms, resilient storage and cyber-security. The trouble is that adding more bandwidth doesn’t equate to increased bandwidth utilisation in the face of the Distance Tax. It’s entirely possible to increase the size of a pipe and to see no noticeable improvement in network performance. It can often prove to be a false economy to repeatedly upgrade network links.
“Conventional wisdom says if we increase bandwidth, application performance will improve. But unless we decrease latency, performance will continue to suffer,” writes Khalid Raza – co-founder and CTO at Viptela – in his article for Network World, ‘Bandwidth alone won’t solve application performance problems.’
David Trossell, CEO and CTO of Bridgeworks, explains: “You can’t change the speed of light, but with artificial intelligence, machine learning and data parallelisation, it is possible to improve bandwidth utilisation by up to 98%, and place DR sites and data centres far apart from each other without penalty.”
Moving large data volumes
In essence, this is WAN Acceleration – sometimes referred to as Data Acceleration. It enables large volumes of data to move at speed over large distances by mitigating latency and packet loss – accelerating data movements, including backups and restores.
Trossell adds that with this technology, it’s possible to do more with any existing bandwidth while the AI helps to make data flows more efficient and faster without impinging on cyber-security. That’s because – unlike WAN Optimisation – it can handle encrypted data, and a zero-touch approach is followed.
Trossell says that, unlike many SD-WANs, the involvement of AI in WAN Acceleration isn’t an afterthought. It is integral, and so it is extremely efficient – helping organisations such as an F1 one team to respond dynamically to changing network conditions.
Organisations often struggle to complete backups within their allotted windows, or to keep geographically separated data centres synchronised, move research data quickly, accurately analyse Big Data or support cloud migration and restore services with confidence after disruption. Whether in e-commerce, military or healthcare scenarios, mitigating latency could turn what is an abstract IT metric into a question of continuity, resilience and more timely decision-making.
Better WAN performance
So, to achieve better WAN performance, IT leaders could do well to consider the following tips:
- Measure the baseline before buying bandwidth. Measure utilisation, transfer-completion times, round-trip latency, packet loss and recovery objectives across representative sites and workloads.
- Prioritise high-impact data flows. Identify backups, replication, cloud migrations and restores that miss operational windows, then set clear performance targets. Within this context, there should be room for air-gapping the most sensitive data – keeping it offline.
- Pilot WAN Acceleration on real paths. Test over production-like distances and loss conditions, comparing throughput and completion times with the same bandwidth. IBM partner OCF conducted a proof of concept with Bridgeworks. It demonstrated that without any WAN acceleration, there was a drop off in network performance, caused by network lag. However, it proved that with acceleration enterprises can mitigate latency and packet loss to meet strict Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO) over long distances without having to pay for expensive, redundant physical bandwidth upgrades.
- Validate security and resilience. Confirm encrypted traffic remains protected, deployment is zero-touch where required, and distant disaster-recovery sites meet recovery objectives. By having the ability to send and receive encrypted data without having to use encryption keys at either end to unlock data for transmission, and then to re-lock it, security is improved. Nobody can view the data, either, when it is encrypted before or after it is sent.
- Monitor, tune and scale. Track latency, loss, bandwidth utilisation and transfer success continuously; expand only where measured gains beat another capacity upgrade. Much of this can be achieved with the use of AI and machine learning. The reduction in human involvement means that IT teams can focus on other strategic activities.
- Use the network you already have: Intelligent utilisation can deliver more immediate value than another infrastructure upgrade. A bit like in business, it’s often quicker and cheaper to get new business from existing customers than it is to find new ones. Similarly, an organisation’s existing network infrastructure may require less investment than a new one.
Lastly, the next phase of any organisation’s infrastructure strategy is not simply about expansion or modernisation for the sake of it, but intelligent utilisation. That means finding ways to do more with less – and faster – without accruing any unnecessary costs. This means finding and use technology to give your own organisation a tax rebate on latency and packet loss to extract more value out of your network infrastructure and your data with more speed and grater resilience than you do at present.
