CTO roundtable: How operators can navigate the road to 2020 and 5G

CTO Interviews

CTOs from across Europe joined Mobile Europe in an industry roundtable to discuss how operators are on the cusp of great change and great opportunity. Graeme Neill reports.

PARTICIPANTS

Bruno Jacobfeuerborn, CTO, Deutsche Telekom Group

Bryn Jones, CTO, Three UK

Liga Krumina, CTO,  Tele2 Latvia

Yogesh Malik, CTO, VEON

Alain Maloberti, Senior Vice President, Orange Labs Networks

Kester Mann, Principal Analyst, CCS Insight

Graeme Neill, Editor, Mobile Europe

Bengt Nordstrom, CEO, Northstream

Tommy Olenius, CTO, DNA

Brendan O’Reilly, CTO, O2 UK

Marc Smith, Group Editor, Mobile Europe and European Communications

Geert Standaert, CTO, Proximus 

Payam Taaghol, CEO, MYCOM OSI

Sascha Zabransky, CTO, Telekom Austria Group

Fernando Zamora, CTO, Másmóvil Group

Radoslav Zlatkov, CTO, VIVACOM


TOs have a very thankless job” was the blunt assessment of Payam Taaghol, CEO of MYCOM OSI, as he introduced the CTO roundtable. Entitled The Road to 2020, it aimed to explore how operators can build a network for the decade ahead, spanning LTE, the Internet of Things, virtualisation and 5G. Taaghol said the executives were caught between the desire to move forward “with fast, evolving technology that needs a lot of investment” and the barriers, financial and otherwise, put in place by management. 

Discussion got under way with CTOs asked how long they considered LTE’s shelf life to be, with 5G on the horizon. Alain Maloberti, Senior Vice President of Orange Labs Networks, said operators still had a considerable amount of space to play in, from carrier aggregation to cellular IoT. He said: “We must continue to extend LTE because in a lot of countries we have not reached the coverage we have with 2G and this is something we need to achieve.”

While he suggested he expected Orange to continue to invest in the technology until the middle of next decade, Liga Krumina, CTO of Tele2 Latvia, said “a lot depends on our customers, their devices and their use-cases, what they will demand or require”, adding she expects serious investment to last at least another five years.

Brendan O’Reilly, CTO of UK operator O2, agreed, adding: “I see 4G especially in the UK being around for a while. I remember listening to a debate a few years ago about when 2G would be done and dusted and I’m not sure we’ve closed that one down yet.”

Bulgaria’s VIVACOM only launched LTE services last year but it already covers 86 percent of the country. Its CTO Radoslav Zlatkov said it is now pushing heavily into LTE compatible devices, with a corresponding effect on its other spectrum holdings. He said: “I believe that in our country 2G is going to die really fast. 5G will depend on the real needs and I think we will still see LTE existing for some time.”

Kester Mann, Principal Analyst at CCS Insight, added: “There are a lot of aspects of LTE that we have only just touched the surface of. I’m thinking gigabit LTE for example. A colleague went to Australia to witness the early Telstra network there. His experiences were absolutely phenomenal in terms of the capabilities of the network and the experience on the device. That’s an area where consumers haven’t even started using it yet.”

Sascha Zabransky, CTO of Telekom Austria Group, took a different tack, arguing LTE will provide high-speed universal coverage and 5G taking more of a regional use-case driven hotspot approach. “I also think there is the financial equation on the table,” he said. “Which technology works best to fulfil the use-case from a technical capability perspective but also from a cost efficiency perspective.”

Tommy Olenius, CTO of Finland’s DNA, said he expected LTE consumption to be 35 times that of 2010 by the end of the decade. He added: “I see that 4G will remain late into the 2020s, because normally new technology when you introduce from scratch, it takes seven years to reach 80 percent penetration.”

Other operators, including Bryn Jones, CTO at Three UK, and Orange’s Maloberti agreed with Olenius’s data growth rates. Jones added: “The data growth curve is showing no sign of abating.”

To meet this ongoing demand, operators are moving to free up spectrum where they can. Yogesh Malik, CTO at VEON, said 3G is the likeliest candidate as its use cases are fading further away. He said: “3G is dead.” He said the challenge is to get the go-ahead to make refarming happen. He added: “It needs a lot of work with government, regulators and education that actually it’s in the interests of the whole economy [to do so]. The more you move up the telco generations, the better it is for the whole country I’d say.”

Orange’s Maloberti suggested refarming isn’t solely a technology challenge but something that requires the input of the entire telco. Referring to a recent refarming exercise in Spain, he said: “You have to balance the investment and the network to put customers on 4G rather than an extension of 3G capacity, then you have to decrease the devices and make sure that you offer proper devices and SIM cards for your customers. 

“This has to be driven by the marketing people in order to make sure that the customers have the good propositions in terms of [products] and in terms of terminals, and then you can execute on the network and concentrate on the extension of the 4G capacity rather than 3G.”

Another area requiring public sector involvement is that of coverage. O2’s O’Reilly said it has taken the operator four years and “a huge amount of investment” to hit 98 percent indoor population coverage in the UK. He said: “The investment for the last two percent is massive because of the amount of infrastructure that is needed. I don’t know about the rest of Europe but in the UK we don’t have a government and regulator that are prepared to allow that type of infrastructure to be built.

“You need a government and regulator that are keen to support it. But even at that point, the amount of investment is huge. So therefore I don’t think there’s a financial case for 100 percent geographic coverage.”

VIVACOM’S Zlatkov disagreed, arguing his company’s late entry into the market, along with the heavy pressure on ARPU and aggressive targets from shareholders meant it had no choice but to push forward towards 100 percent population coverage.

However, Tele2’s Krumina backed O’Reilly’s point, saying that operators can’t prepare for every single eventuality. She said: “Let’s imagine that we have 100 percent coverage but some construction company can build a new block of flats. Then what will happen with your coverage?”

She added: “I believe that networks will never be ready to hit 100 percent coverage because this outside situation changes all the time...We should find the trade off between the cost benefits and ever-changing environment.”

 

The obstacles between here and 5G

Debate turned towards the next generation of telecoms. Bengt Nordstrom, CEO of Northstream, argued a “paradigm shift” was needed in how operators got their hands on spectrum. He said: “The industry has been on a journey since 3G where the concept is to take as much money from the telecom operator as possible through the clever setting up of auctioning schemes. So you’re taking away money. And the regulator is not approving consolidation of the market, at least not in Europe.”

He suggested an alternative would be to license 5G bands to companies willing to invest in networks and build passive infrastructure enabling sharing, as well as freeing up consolidation rules. As things stand, it is too costly to invest and ultimately the entire market will suffer, especially within 5G. He added: “I think the market would be able to invest more if there were three players in any given market and therefore have more level of playing field between the players in the market.”

Bruno Jacobfeuerborn, CTO at Deutsche Telekom Group, also agreed the current economics do not add up and all parts of the industry need to come up with new ways of deploying new technologies. TAG’s Zabransky agreed, but added a further wrinkle was ensuring operators had the right people in place. He said: “I think the biggest challenge that we have is the skills of the people. I’m not only meaning the skills of the people from the vendors; I also mean the skills of the people within our organisations. You could say ‘well, you have always in the past required new skills with 2G, 3G, 4G, and so on’.

“However, my feeling is that with the 5G ecosystem there is much more changing...With 5G things might happen much faster than we were used to in the past. The main challenge that I see is to inspire our people to find a good mixture of experienced and young people for our organisations and to put their thoughts on the table.”

Zabransky said he was downbeat about whether his company has the right skills in place today, adding he was keen to accelerate the intake of new staff and new abilities. He identified open source as a key area where TAG needed to raise its game. He said: “I’m pretty sure that some of our software people have some skills we are not even aware of. I believe there are also a lot of benefits inside open source that we can use to enable totally new use cases from a totally different cost perspective, but I’m not sure we have enough of a critical mass of people that understand how open source communities are working, how we can influence open source communities, because we the telcos are not usually the ones that have driven open source software projects.”

Another area he drew attention to was software defined networks and network functions virtualisation. Zabransky said while there are enough skills to virtualise core networks, he felt there were not enough skills in place to automate processes on top of this.

VIVACOM’s Zlatkov added operators need to be better educated about what technology can do as a whole. Singling out the sales and marketing teams, he said more needs to be done to translate exactly what the new kinds of technology are capable of. He argued start-ups are faster and better at this than operators, at present.

O2’s O’Reilly said while it was easy to get excited about the technological possibilities of 5G, what was more complicated was making it attractive. He said: “Customers, to a greater extent, are not interested in 5G. They want good coverage, consistency, and they want to be able to use their device as and when they want to. And as operators that is our greatest challenge, driving towards that.”

Fernando Zamora, CTO at Másmóvil Group, said, in addition to the regulatory challenges and the need to discover new skills, part of the problem facing operators’ shift towards 5G was internal. He said: “I have been working in networks for 20 years; something I find that we have not been very good at is really being innovative in the way we deploy things. Operators have always been very conservative, thinking ‘if it has not been tested in 10 networks I will think twice before putting it in mine’ or taking a product from a vendor that is not one of the established ones can be difficult to consider. That makes it difficult to bring the starting point for changing the economics to a different level.”

Three’s Jones switched to fibre deployments, arguing the economics of deploying what is seen as the essential backbone of 5G do not add up at present. He said: “The cost of capacity is huge if you start looking at your future trends.” He said attitudes are changing; where cities once saw their street furniture as clad in gold and charged operators a considerable amount for access, they are now approaching telcos with a view to becoming digitally enabled. But he warned: “The current cost of putting fibre in and the infrastructure for 5G, the economics now just won’t work.”

Proximus CTO Geert Standaert said return on investment is set to become an even bigger issue in the years ahead, as operators continue to give away huge amounts of data but with little, if any, knock-on on its ARPU. He said this even applies to golden goose technologies like the Internet of Things. He said: “We were one of the first ones to launch LoRa, but before a company comes to a concept where it is ready to scale, we see it takes ages. They test it, they poke it, then they bring it to their boards etc, but we are already interacting with the company, we put in energy and a massive amount of resources, but there is not yet one penny coming in.”

 

Business transformation and the appeal of failure

MYCOM OSI’s Taaghol said he understood the frustration of operators not being able to gain what they consider a fair return on investment. He said: “Three kids in the back of their garage can come up with a video app that records 15 seconds and deletes it afterwards, with no absolute purpose for any industry whatsoever, and they get floated in the stock market for X billion dollars. Yet we are providing an essential service to entire communities, people will really not be able to function without our service today, it’s fair to say. And we’re unable to monetise the network.”

However, there are similarities; while start-ups work out how to monetise their customer base when they acquire it, operators already have the customer base. He said: “I think we have more valuable data in our networks today that is untapped and that information has much higher value than probably the connectivity.”

He said the possibilities for using that data are “endless”, giving the example of operators getting involved in the government census. “Imagine if we could provide all of this data, give it context, in an anonymous way - I’m not suggesting we sell private information – [then] we could actually extract more value out of the network.”

He said virtualisation could aid this shift as it would give operators the ability to move away from their existing way of life. “For the first time we have the ability to move away from the heavy capex model, which is a big problem for us, where everything we want to do involves huge capex investment with uncertain return on investment and questions and so on to a model where a lot of your cost base is moving to pay as you grow, pay as you use, pay as you generate revenue, and you have a more streamlined cost model with your revenue model in this new world that would allow us to grow.”   

O’Reilly said O2’s use of data goes that into how it builds services.  He said: “What we’re trying to do with O2 in the UK is, when you think about analogue and voice we were making network decisions with the customer in mind. In the digital world you need to be making customer decisions with the network in mind.

“It’s that slight change of focus where instead of making macro-level decisions you’re making customer-driven decisions and using data to help do that.”

Going back to 5G, he said customer-driven decision making could change how networks are built. “Nobody builds a building without water, power or gas,” he said. “The future construction companies should not be building infrastructure which does not have mobile comms at the heart of it, and that I think is the opportunity ahead of 5G. Because you are spreading the cost and the risk and you are making it cheap from the start.

“We all know if you go into a building at the end it costs a fortune. If you do it at the start it’s part of the fabric of the building. That is the opportunity that 5G brings from the infrastructure point of view.”

DNA’s Olenius picked up on the earlier conversation about virtualisation and said DNA was somewhat sceptical towards the technology. He claimed the technology was “not really developed” and also a hard sell internally. He added: “We would rather wait and see. We have invested a lot because we have to grow the capacity hugely. I don’t want to replace all of that investment overnight with some new technology, which is not really bringing any new capabilities for what we are doing.”

However, Tele2’s Krumina argued it was a “logical evolution” to shift towards virtualised architecture. Three’s Jones added: “It comes to looking at the scale of the network that you have to grow to. The first thing you need is the cloud-based technologies enabling you to scale faster. Where we’re putting in capacity from a virtual perspective you can put it in in a few weeks or days or just turn up the scale, whereas a traditional project would take six or nine months.”

It’s not without its challenges, he noted. He said: “I think the real complexity is do you deal with the one-stop shop vendor and get a pre-integrated solution or do you work with a lot of the smaller start-up type companies? When you start looking at building a new overall stack with lots of different components, you’ve got lots of integration issues.”

The challenge is one of risk, where do you put it and who handles it, either a big legacy vendor or smaller start-ups. He said Three chose not to go down the AT&T route of playing, and ultimately struggling, with lots of smaller vendors, opting for the “one-stop shop” approach. He was coy on details, saying not much beyond it being early days for a project that is on track.

Orange has also chosen a more measured approach, with Maloberti saying “we cannot just throw things away”. He said: “What we are doing currently is rolling out virtualised parts when there is a service that needs it. There’s the end services we have launched for enterprises, for instance.”

He added: “I think the big challenge as mentioned is it is a new way to deal with vendors...but I think it is also a completely different way to operate the network to what people are used to. The hardware with the software on top of it, once it is virtualised I think we need to change completely our operational organisation, especially when we go to virtualised RAN, and this is also the challenge that we will face in the next years. So this is why I think this will come step by step and not all at once.”

O’Reilly noted how this shift to virtualisation could transform the balance of power between vendors. “As the world of IT and telco collide, actually the players who are really good at virtualisation are the IT players,” he said. “The IT players that we work with from a traditional IT point of view know how to do cloudification, virtualisation etc. If you’re one of the traditional telco vendors that is a worrying position. So I think that’s an opportunity that 5G brings that maybe we haven’t had before.”

Másmóvil’s Zamora said he still had concerns about compatibility between the different solutions. He said: “When you say it’s 92 percent compatible it’s like saying it’s incompatible with whatever I want. It’s either compatible and it’s clear or it’s not.” Deutsche Telekom Group’s Jacobfeuerborn sounded the warning that virtualisation in itself can’t cure all ills. He said: “I always say to my guys if you try to virtualise chaos you still get chaos.”

 

How important is failure?

The roundtable fell on the day when Cisco announced research that claimed three quarters of all IoT projects ended in failure. VEON’s Malik was unsurprised, saying that too often products are built without the end consumer in mind. Given the wealth of alternatives, consumers can easily migrate from one app to another, leaving companies floundering. 

Másmóvil’s Zamora agreed, saying that examples of televisions that can be switched off by clapping might look good in an advert, but ultimately that was a functionality that wasn’t actually useful. He said: “People working in the gaming industry were doing more complex graphics and then the Wii comes out with a simple interface. It ended up including a lot of people who didn’t even consider going into gaming because of a very simple interface. They found a market for that.

“People come to me showing IoT solutions saying we have the full box for making your smart home. They show a lightbulb and I say ‘ok fine, so you install this one. If it’s broken, where do you go to find another one? If you go down the street to the first cheap shop where you buy anything you need late at night, can you buy one like this? Or something that is compatible?’ No.”

However, O2’s O’Reilly said the 75 percent figure wasn’t the most interesting one from Cisco’s findings. “Because if, of the 75 percent, people tried it once and gave up, that’s a worrying statistic. But if you’re failing fast and turning it into one of the 25 percent who succeed, which is massively high, then that becomes interesting because it means our industry is learning about failing fast and moving on.”