Wireless Operator Strategies

Wireless Operator Strategies provides both a deep and broad perspective of the operator market, combining granular operator-level and market-level data with ecosystem-wide understanding of wireless operator challenges and opportunities.

February 27, 2012 22:42 David Kerr

Day one of MWC was dominated by the usual array of sexy devices with ever increasing feature lists and ever diminishing true differentiation. Today, we saw more of the same with some more color on tablets plus Microsoft and Nokia driving down the Windows Phone specs and price points to potentially enable the next 1B smartphone users.

More significant for me today were the reactions and statements of leading operators. Operator alliances to promote TD-LTE as well as branding RCS under the Joyn moniker as well as significant discussions of privacy issues were all front and center in Barcelona today.

Joyn apps for Android are being shown off at MWC, and in the coming months they will be joined by iOS apps and devices with the capabilities built in.

There is clear consumer demand for enriched messaging and voice services, and Rich Communications provides mobile

network operators with solutions to address these consumer needs.

? Anne Bouverot, GSMA

We also see the industry moving forward on privacy guidelines issues today by the GSMA being a much needed initiative given the tsunami of apps and the inevitable rising tide of opportunities for abuse.

Further evidence of operators and service providers looking to partner to grow the entire mobility pie can be found in the m-payments arena where Vodafone was top of mind with planned global offering partnering with Visa for NFC.

The operator keynote panel which included China Mobile CEO Li Yue, Vodafone CEO Vittorio Colao, and Ralph de la Vega, CEO of AT&T's mobile business as well as Franco Bernabe, CEO of Telecom Italia painted a picture of an industry with significant challenges in declining arpu, escalating investment costs, growing competition from OTT players and of course those pesky regulators.

February 27, 2012 14:02 Phil Kendall


This year’s Mobile World Congress looks likely to be a useful platform for mobile operators to provide updates on their LTE network deployment plans. There have been a couple of announcements from the show floor so far, though some operators put details out last week to avoid the noise. We are expecting LTE launch plans to emerge throughout the week, but here is a recap of what has come over the last 10 days:

  • Deutsche Telekom confirmed an LTE launch by T-Mobile USA in 2013, following the failed AT&T-TMO merger attempt at the end of last year. It is modernizing 37,000 cell sites in 2012/13 as part of this process and through a combination of spectrum refarming and the spectrum dividend it received from AT&T, it will be able to launch with 20 MHz of LTE spectrum (AWS band) in markets covering half of its current HSPA POPs.
  • eAccess will launch an integrated UMTS/LTE 1.7 GHz network in Japan next month, supplied by Ericsson and Huawei.
  • Everything Everywhere hopes to launch LTE services in the UK by the end of 2012, providing it can get regulatory approval to use its 1800 MHz spectrum. The UK is late to auction 800 MHz and 2.6 GHz spectrum, with UK Broadband also looking to step into the void as it trials TD-LTE at 3.5 GHz.
  • Telefonica unveiled its “live test” LTE network (pdf) covering 5 square km of central Barcelona as part of its domestic LTE build. Using Alcatel-Lucent’s lightRadio technology, the companies are positioning this as the world’s 'smartest' 4G network. The network consists of conventional base stations and smaller metro cells, all operating in the same 2.6 GHz frequency band with no interference. It supports speeds up to 100 Mbps downlink and 40-60 Mbps uplink with the partners suggesting each cell can support 30 simultaneous users at 30 Mbps.
  • Movicel will launch an LTE network in Angola by the end of May 2012. The network, supplied by Huawei, is the first 1800 MHz FDD LTE network in Africa.

Watch this space for an update later in the week on further launch announcements.


January 10, 2012 15:48 Phil Kendall

OnStar outlined the latest fruits of its collaboration with Verizon Wireless at CES this week. It’s showcasing a Chevrolet Volt research vehicle connected via VZW’s 4G LTE network to deliver a range of content and enhanced communications services – information services, streaming of multimedia content, video-calling, etc. VZW has been promoting LTE to the automotive industry as a way to future-proof their connectivity strategy and support tomorrow's high bandwidth use cases, and this message seems to be hitting home.

OnStar 4G LTE Research Vehicle 2012

The collaboration makes sense on a number of levels:

Firstly, LTE in general can offer a robust (and fast) enough connection to deliver these services in combination. Watching a YouTube video on its own is not a problem for many networks, but simultaneous sessions across the front and rear seat (video content and calling, online gaming, browsing, etc.) really do need to look at LTE as the starting point – a must-have rather than a nice-to-have.

Secondly, VZW’s network is built on 700MHz spectrum (as a starting point) so offers good geographical coverage. Mobile broadband networks have generally been about delivering massive capacity into urban environments. I visited Softbank’s new LTE network in Tokyo last week – built in 2.5GHz spectrum the cell sites were 50-200 meters apart in order to saturate the streets with high-speed signals, delivering 40-50Mbps to multiple devices in our demo bus. But that network won't extend beyond city limits and the automotive industry needs a high bandwidth network which can offer much wider coverage - the 700MHz LTE network is the only real option today.

Finally, LTE is a reality today in the US market and is here to stay. The US is significantly ahead of the rest of the world in building mobile broadband networks in the lower spectrum bands offering the coverage the automotive industry requires. VZW’s network now covers more than 200 million people in 190 markets (ok, I know I’ve just said there is more to this than urban coverage) and will cover its entire 3G footprint with 4G by the end of 2013. So the US automotive industry, more than any other, can develop LTE services now which will still be supported come the end of their product lifecycles.

The next challenge for the automotive industry will be data pricing. Bundling a year's service with new vehicles could get very expensive (for the OEM) very quickly with these multimedia applications, so new pricing models will be key. Our own survey work has shown that consumers would prefer a single shared data plan to connect their array of 3G/4G-enabled devices, rather than one plan per device. That is still very much a work in progress in the US market, but a necessary requirement if we are going to see the automotive and CE industries enjoy the full benefits of LTE.

December 29, 2011 03:26 suerudd

On December19th. AT&T - discouraged by massive DoJ and FCC opposition - ended its bid to add capacity with the acquisition of T-Mobile USA. This leaves AT&T hunting for alternative ways to acquire sufficient spectrum, even though it was finally able to complete the purchase of Qualcomm’s 700MHz Spectrum on December 27th.

But where does this leave T-Mobile USA?

On December 20th. Deutsche Telekom (DT) CEO Rene Obermann speaking to his own corporate blogger suggested that T-Mobile USA would need to move to LTE technology eventually; and would also require additional spectrum. He made clear however, that the $3 Billion cash settlement from AT&T would be used initially to reduce DT corporate debt not to further enhance T-Mobile USA’s network that was just upgraded in 2011 to dual carrier HSPA+ and is currently providing sufficient capacity for current customers.

Rumors of T-Mobile USA’s difficulties are exaggerated…

As of end of third quarter of 2011 T-Mobile USA served 33.7 million customers only slightly down from 33.8 million at the end of third quarter 2010. 10.1 million of those customers are now using 3G/4G smartphones up 40% from a year earlier. OIBDA margin was 31% in the third quarter of 2011, up from 28% in the third quarter of 2010 due largely to lower losses from equipment subsidies that were reduced by the launch of T-Mobile’s unlimited Value plans. These Value plans allow customers to subscribe to new services without an upfront payment for a bundled handset. These plans have reduced ‘costs per gross add’ and lowered the cost of subscriber retention.

Overall, T-Mobile USA had a slim but maintainable third quarter net income of $332 million on service revenues of $4.67 billion.

This year therefore, T-Mobile survived disruptions to its retail channels including the discontinued retail partnership with Radio Shack; handled the uncertainty of the AT&T acquisition; and held Contract Customer churn down to 2.4%, although Prepaid churn jumped to 7.2% from 6.6% between the second and third quarters.

T-Mobile is maintaining its customer base and making money.

Challenge is Migration to LTE

The challenge for T-Mobile USA is to fund future growth to compete with the three other large players in the US market – AT&T, Verizon and Sprint - as well as ‘no contract’ low cost prepaid regional operators - Leap Wireless, MetroPCS and US Cellular - all of whom are migrating to LTE.

In November T-Mobile announced that its nationwide ‘4G’ (HSPA+ at 21Mbps) network now covered 208 markets across the US reaching more than 200 million POPs.

But how will T-Mobile migrate this network and its GSM PCS customers to LTE without AT&T?

Possible Migration Path to LTE

We have reported that migration to at least a Hybrid HSPA+/LTE network is key to long term profitability. TMoNews the unofficial blog of T-Mobile USA recently described a possible low cost path for T-Mobile’s LTE migration. See: ‘Editorial: Why T-Mobile Should And Will Deploy LTE (The Technical Edition)

At this point in time, T-Mobile is just now refarming PCS spectrum. In a majority of the markets where T-Mobile has both PCS and AWS spectrum with no AWS used, it is quite likely that they will deploy HSPA+ only on PCS because most of the remaining markets only have 10MHz of AWS and 30MHz or more of PCS. AWS will likely be reserved for LTE in these areas.”

“In areas where they’ve got plenty of AWS spectrum and they’re using it for HSPA+, they’ll dedicate about 10MHz of PCS to GSM and 10MHz to HSPA+. If they have 20MHz of PCS or less in a market with lots of AWS, there will be no deployment of HSPA+ on the band, but there will be scaling back of GSM to 10MHz to prepare for LTE. T-Mobile will deploy LTE.”

 “T-Mobile USA has been preparing for LTE on their core network and backend infrastructure for over two years now. … T-Mobile USA has deployed all the necessary components to run IMS on their core network and (has) made a new Wi-Fi Calling solution that uses it. As far as we know, they are the first in the world to commercially deploy IMS for voice, SMS, MMS, and other circuit-switched services. It would be trivial for T-Mobile to change the client software to make it run over LTE or HSPA+. By preparing all the core network and backend infrastructure for LTE this far ahead, their nationwide LTE deployment costs will much lower.” 

Financing the Path to LTE

The key question is whether T-Mobile USA can finance this migration without bringing in new investors. Even if the migration could be done for as little as $6 Billion, additional spectrum will eventually be needed. DT is unlikely to provide that and is being very cautious because of concerns about European market growth and the financial risk of its 40% share of Greek telecom group OTE. DT is also planning to expand in several areas of its European business rather than in the US. 

Nor is DISH Network, who has offered to Partner With T-Mobile likely to be able to finance a T-Mobile upgrade as it focuses on its own rollout.

Meanwhile T-Mobile USA is pursuing business as usual adding ‘no contract’ plans to keep prepaid customers; and offering aggressively priced deals for the Holidays.

And Oh Yes - directly targeting AT&T and iphone4 in its Ads with its attractive T-Mobile girl – Carly Foulkes

December 14, 2011 06:06 suerudd

It has been a busy few weeks for AT&T.



Nov 25th/28th

  • AT&T-deep in talks with Leap Wireless, a second-tier but growing wireless player, to sell it a big piece of T-Mobile’s customer accounts and some of its wireless spectrum…..  AT&T hopes such a deal would placate the Justice Department …… or at least to strengthen AT&T’s hand if it goes to trial.


Dec. 9th

  • US Department of Justice (DoJ) argued before U.S. District Judge Huvelle that since AT&T has pulled its merger application from the FCC, the issue is no longer pressing and there is no need to rush to trial. DoJ asked the judge to delay the trial to an unspecified date in the future.

Dec 12th

  • AT&T has until Jan 12th. to file a report with the court explaining whether it still plans to try to buy T-Mobile.
  • AT&T said that it is considering "whether and how" to proceed with the proposed merger, which needs both FCC and DoJ approval to move forward. If it presents a plan to proceed on Jan. 12th. the pre-trial process will restart on January 18th.


  • U.S. District Court Judge Huvelle puts hold on Sprint, C Spire Suits and any court proceedings until Jan. 18, while AT&T weighs the future of the deal.

While many are saying the AT&T T-Mobile “deal is dead”, and AT&T has clearly irritated the FCC staff, AT&T is now finally taking a much more conciliatory attitude.

Our recent report ‘It's all about Spectrum - AT&T T-Mobile Bandwidth matched by Verizon and Sprint’ describes how Verizon and Sprint are both buying control of significant extra spectrum. However, the Verizon AWS spectrum purchase from SpectrumCo. requires FCC approval; and once it approves that deal the FCC will find it difficult to argue that a merged AT&T T-Mobile would have an excessively ‘dominant spectrum position’.

Once the FCC allows Verizon’s spectrum acquisition - and after Sprint’s takes ‘virtual control’ of Clearwire’s spectrum - AT&T will have a very strong case that its ability to compete will be diminished if it is not allowed to acquire T-Mobile.

A complete reversal of the original case.

AT&T is working hard to complete the deal and is thought to have two teams pursuing two parallel options.

Option 1. AT&T is negotiating to divest sufficient spectrum and assets to satisfy the regulators. See:

AT&T and T-Mobile: Will there be a Spectrum Fire Sale to Escape Department of Justice and Close the Deal?

See Blog: Could AT&T Settlement Catapult Leap and MetroPCS to Top National Status?

Option 2. AT&T is still preparing to fight in court.

  • Verizon’s recent actions may have significantly helped AT&T’s case.
  • If AT&T makes a new proposal on January 12th DoJ may demand that AT&T refile that proposal with FCC before it goes to trial. [Note: AT&T correctly noted that it is normal to resolve DoJ issues before getting FCC approval. We also noted that in April in: ‘AT&T T-Mobile Acquisition: How long will it take to close?’]

AT&T really…really….really… wants the T-Mobile spectrum for capacity growth and is not likely to give up easily.

But, if AT&T loses at trial there are two other likely options:

Option A. “A network-sharing deal between AT&T and T-Mobile could be established, but this would probably not allow AT&T to reuse T-Mobile's AWS spectrum for LTE. A major blow for AT&T.

Option B. Deutsche Telekom is still anxious to complete the sale as it has major European investment plans for the $39 Billion. It might decide to spin out T-Mobile to a Private Equity partnership pending a later sale to a new player like. Google. [Several options for T-Mobile’s future are summarized in AT&T and T-Mobile: Will there be a Spectrum Fire Sale to Escape Department of Justice and Close the Deal?

We should know by January 12th. 2012, if AT&T can come back with a modified proposal that may be acceptable to DoJ, and then the FCC.

Rethinking US Market Structure and Competition in an IP world.

In a recent policy paper Strategy Analytics suggested that the nature of US Mobile Broadband competition has changed and that ‘All-IP’ networks dramatically change both ‘economies of scope and scale’ and the ‘Relevant Geographic Areas’ that determine Competitive Concentration for Anti-Trust purposes.

See: Policy Insight: New Mobile Industry Structure and 'All-IP' Services change AT&T T-Mobile's 'Spectrum Dominance' and Create new 'Challengers'

There may not however, be sufficient time in the current AT&T case to make such a profound change to the traditional 1980s/90s market analysis. Nor to establish rules for the new digital ‘All-IP’ Mobile Broadband Industry.

September 23, 2011 08:22 suerudd

We noted in the Blog of September 19th. that AT&T has begun to explore possible spectrum sales to other operators to gain approval of its acquisition of T-Mobile USA. Two operators were mentioned as AT&T’s first targets – Leap Wireless and MetroPCS.

Is it possible that AT&T could satisfy the Department of Justice (DoJ) by divesting spectrum to just these two companies? To answer this we looked at whether MetroPCS and Leap have current presence or operations in the right target ‘divestiture markets’ to expand cost effectively by adding spectrum and subscribers from AT&T. 

The amended DoJ complaint of September 16th 2011 states that the “…97 CMAs, identified in Appendix B, effectively represent….area(s)  in which the (merger with T-Mobile US) likely would substantially lessen competition for mobile wireless telecommunications services… each constitutes a relevant geographic market under Section 7 of the Clayton Act, 15 U.S.C. § 18.”

The 97 CMAs identified are those where the Herfindahl-Hirschman Index (HHI) is increased by more than 200 points by the AT&T T-Mobile Deal. HHI is calculated by squaring the market share of each firm competing in the market and then summing the resulting numbers. For example, for a market consisting of four firms with shares of 30, 30, 20, and 20 percent, the HHI is 2,600 – Calculated as follows: 302 + 302 + 202 + 202 = 2,600.

Only three CMAs in the Top 100 are not identified by DoJ as an HHI ‘problem’ - Grand Rapids and Flint, MI and Shreveport, LA. Many on the DoJ list however are in the 200+ to 400 HHI increase range that could potentially be diminished significantly by the sale of Spectrum to the third or fourth largest player in each market. 

To see what role MetroPCS and Leap could play we then conducted an analysis of these 97 ‘problem’ Cellular Market Areas (CMAs). In AT&T’s April 21st.2011 filing it listed in Appendices B and C the competitors in all the cities where it operates. The chart below shows that in the 97 CMAs listed by DoJ either Leap or Metro PCS has a presence in all except two - Honolulu, HI and Johnson City-Kingsport-Bristol, TN-VA.

MetroPCS and Leap Wireless presence in the 97 problem CMAs

MetroPCS and Leap are present together in a few CMA’s which might create some competitive bidding in: Philadelphia, PA, Buffalo, Rochester, Syracuse and Albany-Schenectady-Troy NY, Toledo, OH-MI, Lansing-East Lansing and the Upper Peninsula MI, as well as Las Vegas, NV and Tacoma, WA

DoJ’s amended complaint also noted that in “more than half of the (97) CMAs…. AT&T and T-Mobile together would have a greater than 40 percent share. In at least 15 of the CMAs, including major metropolitan markets such as Dallas, Houston, Oklahoma City, Birmingham, Honolulu, and Seattle, the combined firm would have a greater than 50 percent share - i.e., more customers than all the other firms combined.” These are clearly the markets where DoJ is most concerned that the merged entity should have a lower share.

In the 15 markets shown in the chart below the DoJ Post-Merger share estimate is over 50%.

MetroPCS and Leap Presence in 14 of 15 ‘Problem’ markets

Either MetroPCS or Leap is in each of the 14 markets. The only exception is Honolulu, HI where neither is currently present.

AT&T needs to sell significant spectrum, and could potentially accomplish most of the required divestitures with sales to MetroPCS and Leap alone. Provided these two operators can raise sufficient financing – potentially even guaranteed by AT&T - this could catapult these two operators to National Status overnight.

AT&T and T-Mobile’s latest subscriber market shares – as well as those for Sprint Nextel, Verizon Wireless, MetroPCS and Leap Wireless - can be seen at ‘Wireless Operator Performance Benchmarking Q2 2011’.

September 19, 2011 21:23 suerudd

According to two sources speaking to Bloomberg “AT&T is approaching smaller rivals including MetroPCS Communications Inc. (PCS) and Leap Wireless International Inc. (LEAP) to sell spectrum and subscribers” and find a way to gain approval for its acquisition of T-Mobile USA. Sources also said that AT&T “has … reached out to CenturyLink Inc. (CTL), Dish Network Corp. (DISH) and Sprint Nextel Corp. (S) to gauge their interest in buying assets.”

This is the type of strong positive response we predicted in the recent Insight ‘AT&T and T-Mobile: Will there be a Spectrum Fire Sale to Escape Department of Justice and Close the Deal?’ and the earlier blog ‘AT&T Opponents Shift Focus to Challenge Excessive Spectrum Consolidation’. It is likely that AT&T is also reacting to the proposed six month Department of Justice (DoJ) pre-trial schedule and the co-filing by seven states who joined the DoJ lawsuit last week.

Proposed Department of Justice Schedule would go ‘down to the wire.

On September 16, 2011 Department of Justice filed a schedule proposing Monday, March 19 2012 for all steps to be completed in order to be ready for trial. AT&T is anxious to accelerate the process and suggests a completion date of Monday, January 16 2012. This is probably not because AT&T is anxious for a trial but because it still hopes to find a negotiated resolution with both DoJ and FCC as well as the states prior to its March 20, 2012 deadline for enforcing the deal with Deutsche Telekom.

The details of the two proposed Schedules are as follows.

AT&T and DoJ Proposed Schedules

The two parties will meet with US District Court Judge Ellen S. Huvelle on September 21, 2011 to finalize the schedule. AT&T’s proposed schedule is extremely tight for this extensive case.

And the other Parties are Piling on.

Also on September 16 Attorneys General in seven states filed as co-plaintiffs in the DoJ lawsuit. The seven states are led variously by Democrats and Republicans: California, Illinois, Massachusetts, Pennsylvania, Washington, New York, and Ohio. DoJ has had an excellent working relationship with several states in building its case and welcomed their participation.

On the same day Sprint’s attorneys filed motions in Federal Court asking the judge to integrate its case against AT&T in a coordinated proceeding as part of the DOJ complaint.


These opponents’ actions appear to have stimulated AT&T to initiate aggressive bargaining. AT&T and T-Mobile’s latest subscriber market shares – as well as those for Sprint Nextel, Verizon Wireless, MetroPCS and Leap Wireless - can be seen at ‘Wireless Operator Performance Benchmarking Q2 2011’. User shares are available on request to srudd@stratregyanalytics.com or Pkendall@strategyanalytics.com.

May 3, 2011 20:14 suerudd

On April 27th.several competing operators - Cincinnati Bell Wireless, MetroPCS, NTELOS, the Rural Cellular Association (RCA), the Rural Telecommunications Group, and Sprint Nextel Corporation all requested the FCC to consolidate into a single proceeding the upcoming review of the T-Mobile acquisition (11-65) and the earlier Qualcomm MediaFLO Spectrum acquisition filing (11-18).

“In the proposed Qualcomm transaction, AT&T seeks to acquire Qualcomm's six Lower 700 MHz D Block (6 MHz) licenses, which collectively have a nationwide footprint, and five Lower 700 MHz E Block (6 MHz) licenses in five large markets. The T-Mobile transaction includes the acquisition of an additional 50 MHz on average in the same geographic areas covered by the proposed Qualcomm transaction. Obviously, with such an overlap, these two transactions should be examined together, rather than separately.”

In Appendix B. Competitor Chart of AT&T’s April 21st. Filing there are 567 pages of Competitor Charts for 700 MHz, Cellular, and SMR Licensees that show the Qualcomm D & E Block Spectrum alongside the AT&T Consolidated Spectrum. A Selected example for San Francisco, San Mateo and Washington DC is shown below.

As seen in the above example AT&T’s own Appendix B strongly supports the case for examining the joint impact of the two acquisitions.
AT&T may now regret not having waited to settle the Qualcomm acquisition before seeking to acquire T-Mobile. [Note: Appendix A shows the Spectrum to be consolidated between AT&T and T-Mobile.]

So now the FCC may need to review the competitive and market impact of AT&T's aggregation of spectrum in both proposed transactions on nationwide mobile telephony and broadband services.

AT&T may have trouble establishing that it really needs all the additional Spectrum.

RCA President and CEO Steven K. Berry has called AT&T’s acquisitions a ‘spectrum grab’ and said, “It is clear that AT&T is doing everything possible to gain market dominance by making not one but two major spectrum grabs in the proposed Qualcomm and T-Mobile transactions.”

AT&T is now getting increasingly unfavorable publicity – as Broadband DSL Reports notes “Press (Are) Not Buying AT&T Spectrum Claims”. The chart below is being used in multiple publications to exemplify AT&T’s overreach. It is easy visually to add together either the green 3G columns or the blue 4G columns to estimate the consolidated spectrum of AT&T and T-Mobile. It is apparent they both far outstrip Verizon.

Delay may cause problems for AT&T unless it divests properties rapidly.

The clock is running on the AT&T T-Mobile merger since the FCC established the Pleading Cycle for the AT&T T-Mobile Acquisition (WT Docket 11-65) as follows:

Filings                        Due

Petitions To Deny     May 31, 201

Oppositions                June 10, 2011

Replies                        June 20, 2011

 Despite this aggressive FCC timetable, it is increasingly likely that extensive negotiations will be required – both nationally and state by state. AT&T will probably have to divest sufficient properties to ensure competition in each market on a case by case basis to create even stronger competitors than shown on its website.

The agreement with Deutsche Telekom (DT) however, states that AT&T does not have to agree to divestitures and other regulatory conditions that would have an adverse effect greater than USD 7.8 Billion; only half of subscriber or spectrum sales value is counted towards that sum. AT&T has stated that it is prepared to divest up to 40 percent of the number of T-Mobile subscribers or approximately 13.149 million subscribers - based on the stated value of USD 578 per subscriber. But the likely buyers are AT&Ts competitors who have no incentive to move quickly or to pay full price .

But AT&T’s acquisition agreement with DT for T-Mobile USA has a ‘drop dead’ date of March 20, 2012 after which either AT&T or DT can chose to terminate the deal.

If the deal terminates because of a failure to obtain regulatory clearance there is a provision that A&T must pay DT up to USD3 Billion in cash; enter into a roaming agreement; and “transfer to Deutsche Telekom certain wireless AWS spectrum that [AT&T] does not need for its initial LTE roll-out.”

The growing opposition to AT&T may well push the deal to the fall of 2012 as we predicted in an earlier Insight AT&T T-Mobile Acquisition: How long will it take to close?

January 31, 2011 11:55 Phil Kendall

When a company called Sky buys another called The Cloud while a third called O2 plans to outfox them, there is a joke deciphering part of your brain that just goes into overdrive. But I will leave the telecoms puns to the experts.

The UK public WiFi hotspot market is going to change for good as a result of last week’s deals. Smartphones are increasingly the driver of hotspot connections and it is obviously no coincidence that the early movers in the iPhone business, AT&T and O2 UK, have been the more active in the hotspot market – AT&T acquiring Wayport in 2008 and O2 has been central to The Cloud’s resurgence in recent years (and is now going it alone).

We discussed the benefits of these moves for Sky and O2 in a report today. A WiFi footprint in the kind of places you might want to sit down and catch-up on TV is a great asset for the planned Sky Anywhere service, and building a reputation for delivering a better mobile user experience in places of high data traffic should help O2 given that it hasn’t managed to do this with mobile broadband.

So hotspots are great in the smartphone market and will be important for delivering content to larger screen MIDs and tablets. What we didn’t discuss in this report was the future for hotspots as revenue generators in their own right. That’s because there isn’t much. They might be important in stimulating core revenues on smartphone data plans or content-anywhere services, but even the location owners themselves are becoming more interested in using WiFi to drive core revenues (extra cup of coffee or an extra hotel room) than to generate access revenues.

As smartphone use surges, traditional WiFi hotspot players are seeing their metrics move in the opposite direction:

  • KT’s Nespot / olleh WiFi service peaked in 2005 with over 0.5 million subscribers, though has been trending down ever since despite investing in the footprint. Even during 2010, when KT doubled its hotspot location count to 36,000, it has seen subscriber numbers fall 10% to 266,000, and revenues fall 18% (down to 0.9% of KT’s Internet access revenues). It now makes over 7x more revenue from WiBro than from WiFi;

  • iPass recorded a 12% annual decline in its access revenues in 3Q 2010 (which includes WiFi hotspot revenues, hotel Ethernet and also still some dial-up, though the latter is less than 5% of its total). It saw an annual decline of 8.9% in the number of users paying an up-front fee for access to 552,000 users, with the number of users paying for ad-hoc access also falling 14% to 192,000.

Again, as a newly integrated operator, KT is not planning to expand to 100,000 hotspot locations in order to turn around that Nespot business – it’s all about enhancing the mobile broadband user experience and managing costs through traffic offload. And the same can be said for Sky and O2.

December 30, 2010 22:12 suerudd

Skype today launched Video for iPhone, iPod Touch and iPads. This new version of the Skype service application software lets users make and receive video calls from iPhones, iPod touch and iPads, with instant messaging for other Skype users, over both Wi-Fi and AT&T's 3G network.

Was it a test for this iPhone video application that brought down Skype's Video Network Last week? The story going around last week was that a new release for Apple software - possibly the Skype iPhone Video application announced today - had a problem and triggered the Skype server failure when installed first on one and then several Skype 'supernodes'.

But don't blame the Apple software application.

Skype's supernodes act as both offline message (IM/SMS) relays and as Skype's Chief Information Officer noted yesterday "a directory, supporting other Skype clients, helping to establish connections between them and creating local clusters typically of several hundred peer nodes per each supernode."

The initial crashes brought down 25% to 30% of the Skype supernode servers - just before the normal daily peak. This in turn led to traffic overload that created extensive delays in the support servers responsible for offline instant messaging. This resulted in long response delays to some to Skype Windows clients and 20% of these had an old software bug that then caused them to crash.

The official Skype story was released yesterday by Lars Rabbe, Skype's Chief Information Officer, who describes the "snowball" effect that blocked most Skype users for 24 hours on 22nd.- 23rd. December 2010.

"50% of all Skype users globally were running (an older) version of Skype for Windows, and the supernode crashes caused approximately 40% of those clients to fail. These ... included 25–30% of the publicly available 'supernodes', (that) also failed as a result of this problem."

"The failure of 25–30% of supernodes in the P2P network resulted in .. massively increased... load as (supernodes) reconnected to the peer-to-peer cloud... just before our usual daily peak-hour (1000 PST/1800 GMT)". As users tried to reconnect to the system, they generated "traffic to the supernodes that was about 100 times what would normally be expected at that time of day" and overwhelmed the remaining supernodes bringing the whole system to a standstill.

It is interesting that some sources focus blame on Microsoft, not just Skype's network, servers and software, but maybe the problem is more profound.

P2P Server Architecture.

Serious questions need to be asked about a network service architecture that allows:

  • Application software to crash what should be 'carrier class' servers performing network functions
  • P2P software that causes both network and user device based clients to crash as a result of network overload problems
  • Network server problems that spread automatically across a large number of supernodes

Network servers need to be especially resilient and intelligent in how they 'fail-over' in a distributed networking environment; but a robust Service Architecture is always a pre-requisite.

Let other P2P and 'Cloud' service providers beware.

On a positive note Skype brought in massive extra capacity to stabilize the network and was also able to restore Group Video Calling functionality in time for Christmas.

Software Release Deployment

Lars Rabbe also committed to review Skype's "testing processes to determine better ways of detecting and avoiding bugs which could affect the system.". Hopefully this promise includes:

  • 'Old fashioned' regression testing of all old versions of client software
  • Large scale network testing that does not impact live users - especially at peak traffic times!

These are rules that traditional service providers have followed for decades. Perhaps a little more respect for the "old fashioned" network operators and their software release processes is warranted.