Virtualizing your voice network—why bother?

Situation: You have a machine in execution mode. Your machine is constantly handling all requests for voice services, services that in the circuit switching domain gets fewer requests year over year. Why risk anything by changing to new technology? Why would anyone invest in new equipment for a service that gets fewer requests and faces declining revenues?

Do you recognize yourself here? I do for sure, and every time I meet a service provider and start talking about virtual Mobile Switching Centers (MSCs), I get questions that are close to identical with the above paragraph.

Is there another side of the coin? Yes, of course, there is. But again, how valid that side is depends very much on your strategy.

Slim, slimmer, slimmest

Feels familiar? The most common of all strategies today in telecom is cost cutting. So, will virtualizing your network (now, I just don’t talk about the circuit-switching part) lower your run rate? And, more specifically, what will the contribution be from virtualizing the circuit network?

We have worked with some service providers developing processes, tools for quick and easy deployments, and tools for quick decisions.

When it comes to the quick decisions, we audited the virtualization projects we have executed. To be more precise, these projects should rather be described as network consolidation projects. Having said that, I think you all by now understand where I am going by now.

The gains of virtualization and network consolidation

Network consolidation means less nodes, fewer network centers and then, ultimately, reduced cost for O&M. A simple calculation tool that we have used can be downloaded at the end of this blog post. It should be noted that we in the business case, we only investigate cost savings through the freeing up of O&M resources. Areas as footprint reduction and the corresponding impact on rental cost for facilities have not been considered. On the contrary, I do believe that freeing up that space can be a useful tool when looking into revenue growth (as discussed in Break the curve and Voice is not noise).

From a network consolidation view, we made some initial assumptions that ended up being +/-10%, compared to the detailed assumptions made by the finance department. The key hypotheses in the business model (that, of course, are changeable) are:

  • Network opex in relation to total opex: assumed to be 20%.
  • Cost of O&M staff, excluding field maintenance, in % of network opex: assumed to be 15%.
  • Freeing up of O&M resources through network consolidation: assumed to be 40%.

In the cases we have run—since we talk about virtualization and network consolidation—we see that the gains really come into effect when you have more than 10 million active voice subscribers.

What will be your situation? Fill in the yellow labels in the downloadable excel spreadsheet and see! In this general spreadsheet, we have not entered the cost for upgrading software and hardware, nor the support fees. This is something you can get from your supplier.

It should be noted that in all the cases, with more than 10 million subscribers, that we have done this exercise, we have been able (with upgrade and support costs included) to show that the project is cash flow positive within 12 months.


What is the future of voice revenues? In another blog post, Mariarosaria and I talk about re-inventing the voice service; that post, however, is more geared toward VoLTE terminals.

In 2017, there were some 5 billion circuit-switched voice subscriptions. This figure is estimated to be around 3 billion five years later. One can argue, why invest when the amount of circuit switch subscribers decrease by 2 billion in five years?

In 2022, the number of circuit-switched voice subscribers will be of the same magnitude that the total amount of mobile subscribers were in 2007: 3 billion. If you can be cash flow positive within a year, doing both a network consolidation and virtualization and knowing there will be billions of subscribers out there five years from now, then you need to ask yourself—why wait?


Download calculation tool



The Ericsson Blog

Like what you’re reading? Please sign up for email updates on your favorite topics.

Subscribe now

At the Ericsson Blog, we provide insight to make complex ideas on technology, innovation and business simple.