With 203 million customers across Europe and Asia, Norwegian incumbent Telenor is one of the world's biggest mobile service providers. But as its traditional markets become saturated and commoditized, Telenor is exploring a variety of different pricing and service models -- including a move into entirely new fields.
Fresh from Telenor Group (Nasdaq: TELN)'s $360 million acquisition of digital advertising specialist Tapad, head of research Bjørn Taale Sandberg met with Telco Transformation to discuss the impact of 5G and virtualization technologies -- hot topics at last week's Mobile World Congress -- as well as the importance of digital transformation to the operator's future.
Telco Transformation: What plans have you made regarding 5G so far?
Bjørn Taale Sandberg: We haven't announced any concrete rollout or trial plans but we are active and participating in research on 5G and we'll be a part of standardization efforts this year. We think it's important to be a part of the shaping of future standards so that it's not just done by vendors.
TT: Are there any specific regulatory and business needs for European operators when it comes to 5G -- regarding spectrum, for instance?
BTS: That is an interesting question. When you are a pan-European operator you will optimally need the same spectrum in all countries. For us we are spottier in terms of which countries we are in so it's not as big an issue. We want 5G to support the spectrum that will be available and we want it to support both lower frequencies and higher frequencies. When it comes to regulation, one of the big things with 5G is this concept of network slicing -- which means for this vertical there are specific requirements about performance and coverage. And slicing through software control allows us to do that without building separate networks for verticals, which is a bad idea from an operator point of view.
TT: What does that mean for the design of the network itself? Does it make technologies like SDN and NFV more important?
BTS: Definitely. They will be a core part of the solution for 5G and obviously we'll be doing that before 5G is rolled out. Everyone is looking into NFV now to be able to do this network slicing. Then you get into the question of whether the regulatory regime will support that. Our position is that the EU has come down in a fairly balanced way on net neutrality -- they do open up for specialized services, which can have their own requirements. The question is will you be able to define all verticals as specialized? That we don't know yet and there is lots of work that needs to be done.
TT: So net neutrality is actually a bigger deal in the 5G world because of this concept of network slicing?
BTS: Our thinking about 5G is that you should be better able to offer a bespoke quality of service in terms of bandwidth and latencies to specific customers, and our system is designed to allow that without building separate physical networks. It's done through software. If regulation doesn't allow you to do that then you forego an opportunity. In my spare time I build apps for fun and one of them is a network app and it requires some sort of performance. Today I can buy server performance from the cloud service where I host my application, and I can say I want I will this memory and so on, but I can say nothing about the network performance. I have no say over how well it will perform on the phone of the customer. Because of that I have had to spend many hours writing code to cater for bad network conditions. That is a lot of effort. If I could spend €4 euros [$4.40] a month extra on my subscription and know that I have at least 250 Kbit/s of bandwidth to a customer at any time I would save a lot effort. But under current network neutrality current rules I couldn't do that unless it was part of a specialized service that is defined as exempt from the general rules.
TT: Those rules seem to be quite confusing with lots of scope for interpretation. Deutsche Telekom has said it thinks it would be able to charge OTT [over-the-top] players for an assured level of service.
BTS: We don't think we can do that right now but as you say there is still confusion in this area.
TT: It sounds like an opportunity for lawyers?
BTS: Yes, there is a business model there for lawyers!
TT: What use cases with regard to NFV hold the most promise?
BTS: We are researching this and testing it out lab-wise but I don't think any of our businesses have so far committed to wholesale rollout. But it is one of the things we'll need to do -- to "software-ize" the network. It is going to happen. We have a small trial MVNO going in Norway, which we call Lowtel, which has built a core stack using mostly free software and is running a full 3G operator based on that. When it comes to use cases, there is the obvious one of cutting costs in the core. It is also interesting to see new digital services we can play with and how much better we can get at rolling things out quickly. This Lowtel test -- they have built this with something like 12 engineers and it's in operational use with a few customers -- you get a call on your mobile and when you are in range of a PC you get a message about transferring the call to the PC. You can do that seamlessly and it then brings video up at the same time. It's easy to do in software and hard to do the old way. It exemplifies what you can do. So we believe there is an opportunity to revitalize telco services.
TT: How mature are the NFV technologies at the moment.
BTS: Not totally. The upside of software is you can modify it easily and quickly; the downside is there are bugs. That is one reason to be a bit conservative about committing wholesale. You can't afford to have the network out for hours. We had an incident in Norway because the signaling from a foreign operator brought switches down and made it impossible for people to make calls. If the risk of that increases because of virtualization it makes things difficult. So we won't be committing wholesale until we're confident it delivers something like 99.999% reliability. But we don't think we can sit back either. We need to be one of the early movers.
TT: What are the risks of not moving quickly? Is the chief threat from other operators or OTT players?
BTS: Today the hardest competition is still from telcos. Yes, a lot of the time spent on mobile is spent on services other than what we provide and that will continue to be the case. We don't see OTTs as the primary competition anymore. In a way they are role models when it comes to the digital journey. They are partners and companies that can help us get better at competing. We have shifted our rhetoric from talking about them as a threat to talking about them as part of the ecosystem.
TT: Is it getting easier to work with them?
BTS: Telenor has been quite proactive here. We've said earlier we need to partner with these guys rather than do it all ourselves and not learn anything. We try to be a part of the ecosystem and work with them rather than against OTTs. We have built a global backend where all our telcos are integrated to some extent, which makes the customer base available in all our markets with one API. It has made it easier to make Mobile Connect [a GSMA-backed digital authentication service] available. We have our own ID solution we call Connect ID, which we have implemented in most business units -- in all Asian units except India -- and through a single API that allows us to implement the GSMA mobile ID solution in all our markets and that is because of this global backend.
Next page: The digital journey
TT: Where is Telenor on this digital journey?
BTS: We are not done. We have definitely started on that journey and we are going to be investing in digital services and in digitizing the whole telco, which means that instead of having digital on the side of the telco business it's going to be part of the telco business. The way we are organized now is to have a CMO [chief marketing officer] and CTO [chief technology officer] as part of the whole business and they have digital as part of product development initiatives, such as WowBox in Bangladesh, which is basically a portal. All of that is in the CMO organization and he is being measured on the success of these new digital products. It's becoming a part of our DNA. We need to change our way of working -- if not become a software company then become a software-ish company and that means taking best practices from OTTs and startups to speed up the telco. We have started that journey but are far away from completing it.
TT: Is it going to require a big investment to get to where you need to be?
BTS: It will be a big investment but we are saying we need to find the money for that in the business. The task is for the traditional access business, through transformation, to find ways of cutting legacy costs to make room for the new. That is the message.
TT: When you look at OTTs and startups, are you learning about the operational improvements you can make from investing in digital services?
BTS: That is one way. Another is to hire people with the right experience. But we are a company with 33,000 employees now and we're not going to replace all those people so we need to add to expertise and retrain a lot of people. We will need lot of new competence in Telenor and that is competence in machine learning, software and methodology -- how to work differently. Telenor Research, the unit I have, started talking about this a few years ago. We said we could spend all our time and everyone in research could be doing machine learning and we would still have more work than resources in a few years, for sure. It seemed obvious we were going that way. We said the same thing about service design -- we need to find new ways of designing the customer journey to think mobile and digital rather than physical when it comes to serving customers, even in markets like Asia.
TT: How is digital transformation affecting your services strategy when it comes to pricing and the provision of services?
BTS: Here there are different views. One view is that the 5G logic and approach will succeed and we'll be able to launch access products that are differentiated on latency and speed. Why not have your camera upload photos [for a lower fee] at night when the pressure on network might be less. Why should it be against regulation to offer such a service? That is an example. In one line of thinking we'll be able to differentiate basic access products a lot more than is possible today and have two-sided business models and charge an app developer for guaranteed bandwidth. If that becomes possible then telcos have an opportunity to have several tiers of pricing -- much more than today. It won't just be one way to end customers and it will be very diversified. With machine learning we could tailor offers to small groups and individuals based on actual needs. There is lots of opportunity for smarter pricing that gives value both ways.
If that is not possible, then that avenue will be more limited. You will need to succeed through your optimization agenda, by cutting the cost base in your core access business a lot. We'll do that anyway but it will be more important and it will be more important to succeed with new services. Most telcos will place bets in both camps. They will work on costs and ways of pricing data that are more intelligent than today and on getting footprints in new service areas.TT: When you talk about new services do you mean new types of access product?
BTS: Not necessarily. Right now we have typical OTT services such as WowBox in Asia and we are now looking at how machine learning could personalize that more. In Asian markets in particular it's not necessarily divvied up yet and we believe that local content and local solutions are going to be important next to big OTTs. We will have social sites better tailored to Thais, perhaps, than Facebook. There are opportunities to play the OTT game.
TT: Because you have the local presence a Google or Facebook might not have?
BTS: Yes, and because at least in this phase of their development it might not be possible or profitable for them to exploit that, while for someone like us or a startup it could be.
Then we have the digital verticals and we're investing into companies we believe will be important. They are sort of left alone, not being integrated into the telco business, but we will be looking at synergies over time.TT: Do you get rights to intellectual property?
BTS: I don't think that's a big thing for us but it could be. And obviously with the last position we made -- Tapad -- we spent $360 million on an acquisition. That's a move into ad tech for us and we'll run that as a separate business. But probably over time we'll look at ways of using that in our business. That is one example. Another example is financial services. We are the biggest payment provider in Pakistan. We are growing in several other markets with different models for financial services. We are also into classifieds. There, we are in markets where we're not a telco at all, like in Latin America. The other thing is buying into startups and looking to develop them further. If we are successful with a number of digital services we are building, like WowBox, they might be spun off as verticals down the line. In 10 years, maybe the verticals will be Telenor and the telco business will be a vertical.
— Iain Morris, , News Editor, Light Reading, Editor-in-Chief, Telco Transformation