Iliad, Ofcom, RCS-e

Let’s get straight onto this week’s main events in European wireless – a new operator in France, Ofcom adjusting its LTE spectrum auction conditions, and  signs of life for the delayed RCS-e launch in Spain. The first two have been fairly widely covered, of course.

Iliad/Free will tell us again that it is possible to come into a market and disrupt with low prices. Look to Yoigo in Spain, where Telia Sonera is gaining some kind of karmic revenge on the MVNOs (and Hutchison) for wreaking such havoc on its own margins in Scandinavia years ago. This time around, in Spain, T-S is the hunter, and not the hunted. It attacked a similar, tri-partite T1 status quo that exists in France.

Of course, the classic T1 response to the incomer is that by fixing on price, they will only sweep up the sorts of customers that the T1s don’t want in any case. Or, alternatively, the T1’s already have an MVNO or subsidiary brand targeted at that market. The difference in France is that the big three have not seen the possibility of such disruption before, and that the disruptor is a serious looking beast. Although the French market is innovative in services, its three operators have had the game to themselves. The spectrum auction has given the competition authorities a chance to introduce a new player to the game and with this player backed not by a finger-in-the-air “invest as we grow” management team, but by one of your actual billionaires, that new player to the game has a real chance to make something of themselves. Free stands a chance of breaking the “no more new entrants” rule that Bengt Nordstrom, of Northstream, outlined to Mobile Europe last year as an inevitable result of revenue-led auctions.

Not much chance of that in the UK market, where Ofcom is mainly concerned with making sure that when it does get around to auctioning of spectrum, it isn’t immediately dragged into the courts by the unhappy “losing” parties.

Its previous proposal to assure Everything Everywhere some spectrum at the 800Mz digital dividend spectrum was withdrawn. This “sub 1GHz” spectrum is seemingly the most prized as it offers the widest coverage (and best building penetration) profile. With Vodafone and O2 already sitting on stacks of 900MHz spectrum, EE with only 1800MHz spectrum, and Three having no 900MHz or 1800MHz spectrum at all, Three and EE were both keen to make sure that when the new spectrum came around, they damn well got some. Yet following some fierce lobbying, Ofcom announced that it is now minded to remove EE’s guarantee of the 800MHz spectrum. It could be that Ofcom has bought the growing industry view that 1800MHz spectrum is actually quite tidy for LTE. It offers decent coverage, with decent capacity, sitting as it does right between 800Mhz (good coverage, poor for capacity) and 2.6GHz (good for capacity, poor for coverage) spectrum – the other block of spectrum up for grabs for LTE.

EE responded by quickly asking if Ofcom would allow it to refarm its spectrum for 4G use. If it can clear enough space at 1800MHz, therefore, EE wouldn’t even need to sit and wait for an auction, with an implied threat that it could even afford to bid less vigorously in the eventual auction – harming the revenue goals of a cash strapped government, but increasing the chances that the other “big two” walk off with a lot more spectrum for less money. In the long term, EE can’t really allow this to happen, as spectrum is spectrum and the most crucial asset out there. It’s not a game of chess exactly. More like a game of bridge, where bidders try to fix the contract to their advantage, depending on the hand they’ve been dealt. At the moment, EE has bid a couple of no trumps. There will be few passes. Who will be dummy, though?

Finally, some signs of movement on RCS-e, the interesting/ promising/ doomed/ irrelevant (delete as appropriate) specification for interoperable IP goodness such as  group messaging, IM, file transfer and “other stuff”. A Vodafone Spain press release yesterday stated that three main Spanish operators had agreed to launch RCS-e services on selected handsets within the first half of this year. It’s still possible that the operators will make a Mobile World Congress deadline, where an announcement could be given most fanfare, but even that would represent slippage from the promise, given at MWC in 2011, of the launch of RCS-e in at least one market before the end of 2011.

Does it matter that RCS-e is a little late? Are such things probably only of interest to industry nerds? Yes, and no. RCS-e is the template for operators to fight back against the OTT threat, from apps such as Whatsapp to the Google and Apple-led services. It offers interoperability, cross platform support, and in time all this done natively within devices. If it is late, and remember RCS-e is already the trimmed down version of the original proposal, then that is a few more months for the other players to keep growing, another 1-2 years’ phone releases without native support. The other aspect is that RCS-e proponents say it is not just about messaging, it will act as an enabler of other services. The implication being that without it these as yet unseen “other services” don’t happen either. If progress is not seen, the other likelihood is that operators conclude this stuff is too hard, and find other ways to make cross-platform approaches work for themselves. That would be waste of a lot of work, for one thing, but also be another knock to the “interoperability is our strength” message the industry loves to rely on.

What was the last big service success for this telco “interoperability”? SMS? Or you could claim, after a few false starts, MMS? It sounds a strange rallying cry, but telcos need another chalk tick in the column marked “interoperability”. Let’s see if RCS-e can deliver that.

Last weeks news this week:
A week ago, your faithful correspondent trotted off with a shiny new 2012 notebook to hear Mike Flanagan, CTO of network optimisation company Arieso, speak about “hungry handsets”. Arieso had conducted a bit of research into a real, live Tier One European operator’s device profile.

Since then, I have mostly been in bed improvising on a flu profile, having strange hallucinations that included Mobile World Congress being held in my kitchen extension (I don’t have a kitchen extension), and wondering why someone had replaced my medical records with my MWC press accreditation applications (Refused).

So then, apologies for the almost total lack of action on Mobile Europe in that period. Illness and staff holidays have done for us.

Anyway, as only about 450 other sites covered the Arieso news you will no doubt have seen it, but just in case let’s remind ourselves of the main thrust of Arieso’s findings (“last week’s news this week”, that’s the Mobile Europe motto).

Arieso found, in its data mining of this European operator, that users of the iPhone 4S demand three times the data that iPhone 3G users do. Users of the Nexus 1 generate twice the number of data calls on the network compared to iPhone 3G users. The top 1% of users generate 50% of the data demand on the network. The top 0.1% generates 20% of all data demand. Of that top 1%, two thirds are dongle or data modem users, and a third are smartphone users.

This was mainly reported faithfully along these terms. But what, I asked, through clear eyes that betrayed few signs of the incubating infection within, are operators supposed to do with this information? Is a device that generates more data a good thing (higher usage indicates a happier customer and drives data revenues) or a bad thing (increased cost of supporting more traffic)? If 1% of users are generating 50% of the data do you, Mr Operator, view that as a problem (“bandwidth hogs”) or an opportunity to charge the heavy users an appropriate amount, and then look to turn the other 99% into heavy users too, so you can charge them? If most of those “extreme” users are sitting one a laptop dongle, therefore easy enough to work out where they are, isn’t that a nice enough problem to have?

The answers to these questions will depend operator by operator, taking pricing strategies, network utilisation and capacity constraints over the air and in backhaul into account. Flanagan’s view was that with the operator in question, it wants to see users hitting the appropriate usage profiles per device, as that would suggest they are happy users. Someone with an iPhone 4S who is hardly using it may sound profitable, but actually that low usage just as likely to indicate someone who may have a problem, and is thinking of churning.

(Nice analogy here, from Flanagan, that mobile operators cannot operate on the New Year gym membership principle – where a gym signs up dozens of new users in the almost sure knowledge it won’t be seeing them lifting any weights in the near future. With the customer acquisition costs operators support, they can’t afford to kiss off future revenues in that way.)

But Flanagan said that the data did prove one overall trend – that data consumption will indeed increase in line with the more bullish predictions. The logic is that as more and more users get devices that either actually are iPhone 4S or are akin to the 4S, then that increased usage profile will be more widely spread. Result = more data. Roughly three times the data, say, of a typical iPhone 3G, or similar device, user.

Flanagan said that in his opinion it is the automatic cloud-based sync of many the 4S’ services that is driving that increased data on the downlink. As other OS and devices catch up, we will see even more services such as Siri, or auto-syncing content services, driving data usage, he said.

So do device vendors also have a duty to make their devices more sympathetic to network conditions – syncing at quieter times, or in preferable locations? Probably. Does knowing that one device tends to generate n times more data than another device really help? A little, but operators really ought to know this stuff already. Does knowing that 1% of your users are responsible for so much traffic help? A little bit more, especially as you also know that most of them are using mobile broadband more as a fixed wireless access technology for home broadband than as a truly mobile technology. Does flu make you dream weird things? Oh yes.

Keith Dyer
Editor
Mobile Europe

Just over a year ago, Analysys Mason analyst Rupert Wood decided he would swim against the tide and produce a series of reports with titles such as The mobile data explosion is a myth and The mobile data explosion: myths and reality.

Wood’s claim was that industry growth forecasts were wildly overblown, and in particular he cast aspersions on the notion that mobile data volumes would double every year. In particular his comments clashed with the predictions of Cisco’s VNI, upon which so many presentations and models were built. We were taken by his boldness, and conducted this interview with him at the time. We came to this conclusion:

Keith Dyer:
You realise that you are going to necessitate a change in every powerpoint presentation on this topic – the Cisco slide is going to have to go out the window,

Rupert Wood:
Well yes. It’s a myth – the doubling of mobile data traffic every year…

Keith Dyer:
And you’ve also put a hole in an awful lot of business cases which aim to take advantage of the capacity crunch – from equipment vendors to policy management, from data optimisation to billing providers…

Rupert Wood:
As I say, you can draw your own conclusions. It wouldn’t be the first time in the telecommunications industry that traffic forecasts have been wrong.

And now, a year later, Wood is back and he’s feeling vindicated. A note from Analysys Mason, referring to last year’s reports, states:

“These papers had a mixed reception. Some vendors in particular were outraged, and others’ reaction was incredulity at prognoses so out of kilter with received industry opinion, and in particular with that of the de facto gold standard in this area, Cisco VNI. Privately, though, some European mobile operators told us we were more or less right

So it is reassuring – for us at least – to hear those two key points recently confirmed by Vodafone (presentation by Richard Feasey to IIC regulatory conference, Washington, December 2011).
Vodafone stated that its annualised European revenue growth in data was 21% over the period July-September 2011, whereas volume growth was just 19% over the same period. This means the value of a GByte is actually increasing.

Back in November 2010 we estimated European mobile data traffic was growing at an annualised rate of 35%. We subsequently revised this upwards, in part because we thought we had underestimated the effects of seasonality on data consumption.

At the same time in 2010 Analysys Mason also wrote:
Myth: The value of transporting a byte is collapsing?
Reality: This was emphatically true up until the end of 2009, but the rate of decline has slowed recently. If operators manage to preserve the premium for small-screen data, the value of transporting a byte of data could actually rise as the mix of large-screen and small-screen becomes more beneficial to the operator.

In fact we thought – perhaps a little cynically – that mobile network operators’ sudden enthusiasm for tiered handset data pricing in late 2010 had more to do with the desire to hang on to the handset data premium than with networks falling over under the weight of data.

Vodafone of course is just one case, and despite its size, is not generally the largest data carrier in its 13 European markets. But it is not alone. The steady trickle of information that comes back to us from the players that really matter in the industry, those at the RAN-face, has mostly – though not universally – tended to confirm what we originally thought.”

So what are the main conclusions Wood and Analysys Mason now draw from this?  Here’s how the statement concludes.

• The idea that mobile operators are the passive victims of some demand force of nature is just muddle-headed. Open-ended demand forecasts that do not take into account operators’ ability to finance data networks are meaningless. And so are forecasts that simply assume operators will have to bow to consumer pressure for flat-rate pricing. Pricing is actually a very effective lever.

• Fixed broadband soaks up a huge share of small- and mid-screen wireless device data. Our own primary-research consumer surveys bear this out.  In markets where there is widespread fixed broadband, the reality is that smartphones and tablets are essentially fixed broadband devices with additional mobile functionality.

• Long-term demand for bandwidth-intensive data in locations where the fixed/Wi-Fi default is not available is not great.

• Most mobile operators should be reasonably happy that fixed operators are bearing this excess traffic for the time being. There is really no compelling reason for these operators to attempt to service this traffic unless they can come up with a genuinely compelling 4G FMS product priced at a similar but competitive level to basic fixed broadband double-plays. And it is this that is at the core of Vodafone’s LTE strategy: not a response to a capacity crunch that may not even exist, but as a platform for next-generation broadband services.

In fact, although Wood’s support for his view is still mainly hidden – outside of those Vodafone numbers – we are also now seeing an increasing number of voices pipe up that mobile data demand is not on the predicted x2 every year curve. Indeed, even the assumption that data revenues, and especially profits, cannot keep up with the cost of transport, will come under increasing challenge – especially following those Vodafone results, in which the operator stated that data revenues were now aligned with costs. So is Vodafone an outlier amongst the Tier One operators, or will we see similar results from other operators come through in 2012? Wherever the truth lies, it is good to have analysts that question the orthodoxy, especially if they can back their questions up with actual data.

So, this is the final newsletter of the year from Mobile Europe. I’d like to thank you all for all the feedback, emails and messages over the past months. We’ve had a great year here at Mobile Europe, and look forward to the next one.

Keith Dyer
Editor
Mobile Europe

European Communications launches customer experience survey

As network operators continue their evolution to being fully customer facing organisations, there is still debate around what customer experience actually means from a telco perspective, what it should entail, how it should be implemented and who should lead it. To shed some light on this, Mobile Europe’s stable mate European Communications is undertaking a customer experience survey to assess the views of senior telco managers around the world. Click here to take the survey.

The results will be revealed at “Understanding Customer Experience”, a special seminar sponsored by Oracle, Amdocs and Alcatel-Lucent. Please do consider both completing the survey and attending the seminar.

Two webinars are now available to view. Both were successfully held this week, with hundreds of attendees joing from across the globe. They are:

Designing high capacity networks in stadiums: challenges & best practices

Making Wi-Fi “Carrier Grade” to Support Cellular Offload