Betting on shares

Network sharing_image

Bearing Fruit: Operators Are Increasingly Looking To Network Sharing To Cut Costs

There has been a good deal of debate of late as to whether or not the commercial arrival of LTE will herald the return of the network to the status of key competitive differentiator. During the past decade carrier psychology evolved and the network, in developed markets at least, began to take second place to other elements of the offering—brand, market positioning, content and service mix—in the efforts of the operators to distinguish themselves from their peers.

The big fear for operators was—and is—that they might end up relegated to the role of transport provider. And the network is the huge, expensive, physical embodiment of that undesirable outcome.

More recently, though, we’ve seen evidence of a shift back to a network-centric view of the world among operators. When Nordic player TeliaSonera became the first operator to launch a commercial LTE offering in Stockholm and Oslo late last year, it was in a bid to demonstrate prowess and leadership in the network, said Håkan Dahlström, the firm’s president of mobility services. “LTE gives us the opportunity to give our customers high quality access, and to really prove to our customers that going with TeliaSonera is a future-proven choice,” he told MCI.

A couple of months before this conversation, Matthew Key, head of Telefónica Europe described the network as the “crown jewels of the customer experience” while discussing the progress of the passive network sharing arrangement the firm has in place with Vodafone. He was at pains to point out that the passive share was an economic necessity, and not a value judgement on the competitive collateral of the network.

If carriers are beginning to look to the network as a differentiator once more, then the timing coincides with a growing enthusiasm for something which has historically been cited as evidence that the network was becoming commoditised; network sharing. Up for discussion is not the kind of passive share Key was talking about, which involves collaborating on land and towers in a bid to bypass the huge cost and logistical headaches of site acquisition. Instead the focus is on a rather more literal interpretation of the phrase; multiple operators using a single radio access network to deliver separate commercial services.

There are a number of reasons for this and, as always, money sits right at the top. From next year LTE deployments are expected to begin in earnest and the costs of deploying a new network, rather than another round of software upgrades, are looking ominous to many operators. If those costs can be shared with a partner, and a network jointly deployed, it should mean quicker deployment and quicker coverage spread.

LTE deployment cannot be viewed in isolation from the capacity crunch that has been the industry’s defining recent trend and network sharing is increasingly viewed as an effective way of beefing up capacity in an economic and efficient way. The problem is summed up unsentimentally by Neil Coleman, product management director at Actix: “This nice, comfortable model—get a bit of spectrum, sell a bit of voice—has been blown to pieces by mobile data. Now it’s all about desperately scrambling to get capacity into the network.”

The other reason that network sharing has earned greater prominence recently is the fact that the transformational project that has seen the networks of 3UK and T-Mobile UK painstakingly combined in a deep RAN share is finally nearing its conclusion. The two firms set up Mobile Broadband Network Ltd (MBNL) in December 2007 to oversee the merging of their two networks and in just a matter of weeks—a few months short of three years since launch—MBNL will have succeeded in creating a joint network comprising 12,400 active shared sites.

Those involved in the project describe it as monumental, a world first. They claim it more than answers the sceptics who said such a transformation—an order of magnitude more complex than a shared deployment conceived as such from the outset—could not be done. And they fully expect other carriers to look to implement similar projects of their own, where achievable.

Not everyone’s gung-ho, though, and a good number of carriers remain focused less on what they might gain through an active network share than what they might lose. Perhaps the most often cited fear surrounds network integrity. “Carriers worry about how much information leakage there will be from one operator to another,” says Sharad Sharma, a consultant at Coleago Consulting, a specialist telecoms consultancy that has worked on a number of network sharing deals.

“If you’re sharing the RAN you might be worried about the competitive advantage of your marketing data, or the volume of traffic you’re generating; that sort of confidential information,” Sharma says. Another potential concern is the level of independence that sharing carriers retain should they want to launch differentiating services. And then there’s the issue of the network as competitive advantage.

These fears have to be weighed against the benefits, say the advocates of network sharing. When the MBNL project was first announced, the parent firms said they would be looking to reap savings of £2bn over ten years. Chris Woodland, an associate partner at KPMG, who has worked on “some of Europe’s leading RAN share deals” backs that number.

“You can clear that sort of value even after you’ve included integration and restructuring costs and decommissioning costs. It’s a good one third of RAN operating costs per carrier,” he says. “And this is before any future capex spend you might have to make. There are quite a lot of avoided costs with network sharing.”

Woodland adds that the network performance improves in terms of capacity and coverage, creating a better customer offering and that there are environmental benefits derived from the reduced physical footprint of the network and the reduction in power requirements. “When you look at that in the round,” he says, “you’ve got reduced cost, improved customer service and an environmental benefit. It begins to look like a no-brainer.”

It is not the case that any two carriers within a given market will make successful network sharing partners and there are more than a few instances where projects have been abandoned

But while the likes of Sharma and Woodland are in favour of network sharing as a strategy, they—along with those involved in the MBNL project, and anyone else you  »
care to ask—stress just how complex the creation of a network share can be. If it’s not done properly, they say, it could easily turn very sour indeed.

Bradley Mead is vice president for services at Ericsson UK, the outsourcing partner for MBNL, and has been closely involved in the project since the outsourcing deal was inked in November 2008. Ericsson also has a wider managed services deal with 3UK. “You do have to have some empathy for the risk involved in this sort of activity,” Mead says. “It’s a complex transition, even though it’s planned activity. I can understand why operators are nervous about taking that risk in the highly competitive environment of today; it’s something you shouldn’t enter into lightly.”

A network share is a relationship and, like any relationship, it requires compatibility, commitment and effort; it can very easily go off the rails. Wtih this in mind, it is impossible to attach too much significance, say those with experience, to the process of finding a suitable partner. It is not the case that any two carriers within a given market will make successful network sharing partners and there are more than a few instances where projects have been abandoned.

“The operators have to want to work together,” says Graham Payne, managing director of MBNL, the man to whom responsibility for the success of the T-Mobile/3UK network sharing project ultimately falls. “It’s easy to look at where you’re getting a leg up, and look at where the other partner is getting a leg up and start to feel jealous—and then it will fail.

“You have to go into it with parent companies who are confident that the benefit they get is good enough for them and who understand that the other partner has to get a benefit as well. So you can’t be spiteful about things like this, you have to be absolutely willing to work together on the network while remaining competitors in the market,” he says.

Such an understanding has been in place between 3UK and T-Mobile since the start of the contract, Payne says, although he concedes that during the negotiations prior to the contract being signed, there were some “heated times”.

Any project of this scale will throw up surprises and require a certain amount of adjustment, and perhaps even improvisation. But as far as is possible, operational details and commercial principals have to be nailed down during the negotiation phase, says KPMG’s Chris Woodland. “What happens if one party’s technological roadmap is different from the other party’s?” he says. “What happens when one operator wants to launch a service that the other operator doesn’t want to launch?” It is vital that network sharing partners try to pre-empt any challenges that are inherent in the project ahead of the project beginning.

The decision also has to be taken as to how the new shared network will be administered. There seems to be almost universal agreement that the creation of a third party organisation like MBNL is the most satisfactory way to go about it. A third party organisation, even if it is created from teams that come from each of the parent companies, can develop an identity of its own, and be seen as independent of the parents in terms of the job it has to do; create and run a network that serves both owners.

There is an alternative to the joint venture, says Mark Nield, head of business transformation for Western Europe at Nokia Siemens Networks, giving an indication as to future business models the vendor might be chasing. “There is another way to do this, which is an organisation that is at arm’s length from the parents. We believe that a consortium of a vendor and a towers partners could provide the network entirely independent of the network operators.”

Vendors have historically been quite wary of speaking about such direct network ownership but, even in the case of an MBNL-style joint venture, outsourcing to a managed services vendor seems a logical extension of the network sharing business model. Some suggest that the vendor which is most »
prominent in the parents’ networks should be the likely choice for managed services partner but not everyone agrees. MBNL uses
Nokia Siemens kit, for example, but selected Ericsson as the managed services provider.

What does seem clear is that, while a multi-vendor environment in the network need not be a deal-breaker for network sharing projects, a single vendor environment makes things a lot simpler. When MBNL began its project, says Graham Payne, 3UK had NEC kit in place. This was swapped out for Nokia Siemens Networks, partly because of NSN’s superior roadmap, he says, partly because NSN made its pricing attractive and partly because Payne wanted a single vendor network.

“We managed to negotiate a good cost,” he says. “But more importantly, when you’re looking at having two vendors in a network, you’ve got two lots of software, two lots of software testing, two lots of contracts to manage. Moving to one vendor paid back within about twelve months.”

Before such decisions can be made, however, partnering networks need to find out exactly what they’ve got to put into the sharing pot. Many operators, says Coleago’s Sharad Sharma, don’t have a definitive record of their full range of network assets. Securing that information can be a substantial challenge in itself, he says. In one project he worked on, he says, “it took almost five months just to acquire that information, even though the company had done an asset management exercise in the past.”

While all of this is going on, and as operators are trying to ensure that their individuality and integrity will remain after they have combined their networks, they also have to think about the most important part of the equation; their customers. Merging two mature mobile networks is a hugely disruptive task and it is pretty much guaranteed that some of the pain is going to be passed on to the end users. Sites need to be taken down for hours at a time and the service they receive is certain to be affected in a negative way before the planned outcome of an improved network can be realised.

“This is the area that presents the biggest challenge,” says Ericsson’s Bradley Mead. “When you’re changing things like this there is down time associated with it and with MBNL the focus has been on minimising that, by doing the work in periods of low traffic, or out of hours,” he says.

Merging two mature mobile networks is a hugely disruptive task and it is guaranteed that some of the pain is going to be passed on to the end users

Both 3UK and T-Mobile embarked on carefully managed customer service programmes ahead of the project in a bid to explain to their users what was happening, why it was happening and what the likely disruption would be. Watching from MBNL, Graham Payne suggests that overall customer service improved as a result of the project as the carriers learned how to manage such complex communications.

For its part, MBNL had to put in place “totally robust performance management processes,” Payne says. “As soon as we did a change on the network we made sure that there weren’t human errors. Immediately after the change the network statistics and KPIS are checked and test calls are made. Then that site or change is monitored intensively over the next r24 hours and checked for seven days.” It was a skill that MBNL improved as the project went on, Payne says.

While the parent operators were trying to manage the expectations of their customers, Payne’s team were focused on the nuts and bolts. He describes MBNL as the most monumental project of his career and, with that in mind, says that it is difficult to pick out the single biggest technical challenge that he faced. But forced to choose he suggests that getting the radio network controllers integrated to the parent companies’ core networks was the biggest headache.

The original plan was to do all of the cell sites on a piecemeal basis, he says, but that proved unworkable. With the different sites, the different RNCs and the different data centres it was “like spaghetti junction out there,” he says. The network consolidation was taking time to come through, he says and the breakthrough came when the team had all of the RNCs integrated, and all of the sites onto the correct RNC.

But market circumstances weren’t helping, he says. “The thing that made it even more complicated was the traffic growth in mobile broadband that was happening in parallel with us doing this consolidation,” he says. “That just trebled the complexity of the task.”

At times the project seemed like it might be insurmountable but the end is now almost within touching distance. Those involved are claiming much improved network coverage and performance of the combined network and offer reassurances that the kind of concerns that might be worrying operators thinking about embarking on something similar are unnecessary.

Despite the depth of their share, T-Mobile and 3UK actually have different call drop rates, different set up rates and network availability. And under the contract that was established they each have the right and ability to enter into unilateral network deployments. If they do, of course, they immediately lose the benefits of the sharing programme they have worked so hard to establish. Now, says Payne, they share their plans and are more likely to work on shared deployments than to go it alone.

The industry’s reaction to the MBNL project will be interesting to watch. And it will depend very much on the performance of the network in the months and years ahead. If successful it may well stand as a landmark project, one that many other carriers seek to replicate. There is little time for the teams that worked on it to sit back and admire their handiwork, though. The merger of T-Mobile and Orange in the UK to form Everything Everywhere means that another, possibly even larger scale transformation looms up ahead.

As for the future of network sharing, NSN’s Mark Neild reports that he sees pressure mounting for sharing projects in back office functions like billing systems, partly driven by outsouring offerings from the likes of IBM. And one of his customers recently asked NSN to look at the opportunities for sharing to extend into the core network, he says.

MBNL’s Graham Payne says he sees no reason why network sharing shouldn’t progress to the stage where we have single network markets for the majority of the network. Operators could differentiate on in-building solutions rather than on the macro network, he says, although others suggest that two network markets are more likely.

He offers one final word of caution, though, which is that network sharing is a one-way street. Once two operators have set off on their journey they are stuck together, he says, with no opportunities for a quickie divorce. “It’s an awfully big step and, when you take it, you’re absolutely locked to one another. Because unbundling it is more difficult than merging it together.” Marry in haste, if you like, and you’ll repent at leisure.