South Africa has spent years treating fibre as the broadband finish line and wireless as a temporary arrangement until the trench reaches the gate. That hierarchy is now out of date.
Modern fixed wireless can now deliver the speed, capacity and reliability to be the first choice rather than the consolation prize, and newly released spectrum will only sharpen that case further. In the right circumstances, wireless can match or even outperform fibre and reach customers whom fibre may never serve economically.
Don’t get me wrong, fibre remains an excellent technology. Where customers are packed together and ducts, poles or manholes are already in place, it is usually the obvious choice.
But the economics change quickly across farms, scattered settlements, small towns and businesses on the wrong side of an expensive trench – in other words, across the majority of our underserviced communities.
New spectrum changes the ballgame
In May 2026, ICASA gazetted final regulations for innovation spectrum covering two distinct bands: the lower 6GHz band, from 5.925GHz to 6.425GHz, and the 3.8GHz to 4.2GHz band commonly associated with N77 private 5G.
The lower 6GHz allocation gives wireless operators access to 500MHz of comparatively clean, licence-exempt spectrum under a shared framework. N77 provides a route to licensed, dynamically managed spectrum for private 5G and fixed wireless networks. Together, they give providers considerably more room to build high-capacity services.
The regulations still require a separate commencement notice, so nobody should expect 900MHz of new capacity to arrive by courier tomorrow. Even so, the framework is a significant achievement after years of trials, submissions and technical work by ICASA, the CSIR and industry stakeholders.
Regulators do not often receive applause from the industries they regulate, but credit is due here. South Africa has moved from discussing dynamic spectrum access to creating a workable framework while several better-resourced markets are still wrestling with the same problem.
Importantly, the change is not only about fibre versus wireless: it’s also wireless versus wireless. Mobile operators have long sold fixed LTE and 5G routers using spectrum acquired at considerable cost. N77 will allow smaller wireless providers to use similar underlying technology without a multi-million rand investment. The spectrum will be shared rather than exclusively auctioned, but the service will be familiar: a router in the home or office connected wirelessly to a nearby site.
While packages will still differ on uncapped use, fair-usage policies and throttling, that debate can wait. The important point is that the technical gap has narrowed sharply, and wireless services are now in an unassailable ascendency.
Choose the connection that fits the job
This is not an obituary for fibre, rather a case against treating one access technology as the answer to every geography and every customer. In a dense business park with existing ducts, fibre makes sense. If every building is visible from one high point but reaching it requires fresh trenching, wireless deserves to be considered first.
Wireless also provides physical diversity from infrastructure in the ground. There’s no escaping the fact that fibre can be dug up. I once watched a TLB continue excavating with about 15 metres of warning tape caught on its bucket, flying behind it like a kite!
A radio link is not immune to failure, of course. But rather than serving as the emergency downgrade businesses tolerated while waiting for fibre repairs, it is now fast and reliable enough to serve as the primary connection or as a proper secondary link.
Then there’s the numbers. No amount of enthusiasm changes the cost of trenching kilometres of road for a handful of customers. Clean spectrum and modern radios can cross that distance with far less civil work, connecting farms, rural communities, lodges, schools and smaller enterprises.
One wireless site can also carry multiple services: business connectivity, CCTV backhaul, agricultural monitoring and IoT traffic. WISPs are becoming general-purpose connectivity providers, no longer just businesses that filled a gap until fibre arrived.
Where to from here?
South Africa is particularly suited to this shift. We never built the depth of fixed copper access found in Europe or North America, which is why many WISPs emerged in the first place. Fibre replaced DSL remarkably quickly where density supported it, and its footprint will continue to grow. But the pickings become leaner with every kilometre.
The remaining broadband market increasingly consists of places where distance defeats density. From a hilltop in rural KwaZulu-Natal, a house several kilometres away may be clearly visible across open country. Running fibre here an engineering project; connecting it by radio is a flick of a switch.
For those customers, the question used to be whether wireless would suffice until fibre arrived. As new spectrum and newfound opportunities for WISPs take effect, the better question is why they should wait for fibre at all.
- Paul Colmer, Executive Member, Wireless Access Providers’ Association (WAPA)
