Speaking Engagements & Private Workshops - Get Dean Bubley to present or chair your event

Need an experienced, provocative & influential telecoms keynote speaker, moderator/chair or workshop facilitator?
To see recent presentations, and discuss Dean Bubley's appearance at a specific event, click here

Sunday, December 18, 2011

2012 Winners and Losers

It's time for the annual merry-go-round of analyst predictions, and so I'll thrown my hat into the ring as usual. Some of these will come as "no surprise" to my regular readers & clients, others represent a bit of a change of stance.

I'm hideously busy before I head off for the holidays, so I'm afraid these are just bullet points. More detail & explanation will crop up here over time, or on my new subscription @DApremium Twitter channel.

  • Telco OTT services & business models This won't surprise many regular readers as it's a theme I've talked about for many years. It's a complex and nuanced area though, and we'll see as many failures as successes, if not more. Nevertheless, trial & experimentation are essential "you have to be in it, to win it"
  • Loyalty business models for telcos - I think much of the future revenue growth opportunity for operators comes from clever pricing and marketing, assisted by network technology where needed. "Top up your account by $10+ by Wednesday and get a free 20MB of data and a discounted movie ticket".
  • Charging/billing related policy engines  - the key enabler of loyalty schemes and other similar offers. The policy "intelligence" will primarily reside in the IT side of the operator, with enforcement on the network side. We'll also see on-device policy clients extending beyond today's simple dashboard / quota apps.
  • Enhanced telephony & SMS services or functions - Making existing 20-100 year old services better should be prioritised above creating new ones. So-called HD voice is a start, but it's shameful that the industry abdicated its responsibility and to let Apple create Visual Voicemail and Siri, or Palm and others developed threaded SMS. Unfortunately, the obsession with standards and interoperability seems to stop the telco industry itself from improving the user experience of basic functions.
  • Non-seamless operator wifi (offload, onload & other models). "Frictionless" will be a particular winner, not "seamless" (ie user will remain in control most or all of the time). A regular topic of mine: see various blog posts and this white paper I wrote on Carrier WiFi.
  • iPads & Kindle Fires as complementary devices - OK, I'll admit I was unduly pessimistic about the impact of the iPad when it first launched. However, I still see it largely as an *extra* device for people rather than a substitute. I still don't buy this "post PC era" rhetoric. Netbooks were extra devices too, sold to people wanting a cheap portable second computing device - and iPads fit in the same category, with added coolness and new use-cases. I see the Fire as doing something similar - it looks like something new and fun, but again is incremental not substitutive.
  • Microsoft & Nokia smartphones - I'll take a flyer on this and say that I expect MicroNokia to be a surprise success, although I think the buyout rumours are bunk. That said, it's critical we see continued developer traction - my main reason for not getting a Lumia at the moment is lack of support for some critical apps for me (BTFon for WiFi, Onavo for roaming data-saving, not sure if the Barclays London bike hire scheme has one yet)
  • Mobile broadband data plans based on speed, location, time, device & user - all of these criteria can be determined accurately and easily (unlike application type). Ever more sophisticated pricing plans will be developed around them, supported by realtime information about network congestion, and perhaps ways to offer "happy hours" or similar.
  • On-device network-intelligence for connectivity, policy & control (including both telco-driven innovations & more user-centric apps). Another theme of mine recently: trying to manage the network just from the core won't work - there needs to be intelligence out at the real edge to understand what's going on. Trying to guess about "video optimisation" when you don't even know if the user is watching a foreground player, or downloading in a hidden window makes no sense, for example. To know, you need footprint on the phone. Same deal with managing WiFi access and maybe policy in accordance with user wishes & expectations
  • Consent-based mobile video delivery eg based on Operator CDNs, adaptive bitrate streaming &  close cooperation with video publishers. Where the operator works directly with a content company, they get "consent" to reformat or otherwise manage a video stream. That is going to be much more acceptable than trying to use boxes in the network to "optimise" it without permission or transparency. I wrote about the idea here (and comments) and I spoke about this at a conference recently - watch for an upcoming white paper.
  • Wholesale 3G / 4G networks - yes, LightSquared has problems with GPS. But the general model of wholesale-centric LTE networks is a very strong one. Depending on the market, some will see disruptor new entrants, some will see government intervention, and others may see existing tier-2/3 operators opt for structural separation.
  • Freemium Twitter / LinkedIn business models for analysts, consultants & other professional services companies. Sooner or later, someone clever is going to find a way to make money out of Twitter *coughs modestly*


  • RCSe - It was dead. It's now still dead, but shambling around like a zombie. I'm hoping to publish a report shaped like a wooden stake during 2012.
  • NFC payments - classic case of a solution looking for a problem, with the added bonus of squabbling over who controls the wreckage. Massively overhpyed.
  • QR codes - They're everywhere it seems. But have you ever actually seen somebody use one in public? Just give me the URL: I'm not going to "engage" with your marketing until *I* decide to.
  • LTE in Europe - Spectrum, devices, voice, business case. Oh and flattening demand for data as well.
  • SMS revenues - they've been a sitting duck for years. 20 years ago, sending 160 characters of text was quite a feat. Now, not so much. I expect massive price erosion (even within bundles) as a response to the continued viral adoption of iMessage, BBM, WhatsApp & whatever 2012 brings us.
  • Cellular-enabled tablets - Poor fit of subscription "data plan" model with computing devices that tend to get used in unpredictable and ad-hoc fashion, often in places with WiFi anyway. And in the idea that you'd be happy to let an operator policy-manage your computing/Internet experience, and you've got a recipe for disaster. I told you so, 3 years ago.
  • Telco control of user identity - I think the idea of SIM=identity will break down irretrievably in the next 24 months. Operators *do* have a role to play in ID and authentication, but it needs to be much more nuanced and user-centric. Some of the stuff that my associates at Telco 2.0 are doing with the World Economic Forum is interesting, but the industry still has too much of an arrogant belief that it can "own" more customer knowledge than it will get in practice.
  • App-based policy control & charging - Another familiar theme of mine. The network doesn't see apps the way that users do. HTML5 makes the problem even worse - how do you charge for "Facebook" when there's 800m different versions, created on the fly in the server?
  • Eurozone-exposed telecom companies - I'm not an economist, but delving into the topic with Telco 2.0 for a recent paper made me realise some of the problems that next year might bring. Hopefully I won't be writing the 2013 predictions from a cave, looking at the smouldering ruins of civilisation. All things considered though, I'm quite glad to be living and working this side of the Channel, despite the Monty-Python inspired taunting from our friends in Paris.
  • ANDSF, IFOM, I-WLAN & seamless wifi offload - There's some good stuff in ANDSF (eg advising the user which WiFi access point is best), but it's a bit buried in the general belief that operators can control users' WiFi overall, forcing connectivity and enforcing policy. They can't, expect in a few corner-case instances. Attempting over-control will backfire: telcos can either control 30% of user WiFi, or zero. I suspect they'll aim for 100%, and the push-back will mean they'll get zero.
  • Operator API balance-of-payments vs. Internet companies - Telcos are still pushing "capability exposure" via initiatives like OneAPI and Telefonica's commendable BlueVia project. But they're still not doing the right APIs in many cases (eg "does the user have a suitable data plan?", or "does the user have good coverage & network speeds?"). Meanwhile, Google, Facebook and others are starting to charge for bulk access to *their* APIs: telcos are often major users. Result: the net API balance of payments won't be pretty.
  • Full mobile network outsourcing to "under the floor" players - another theme I've talked about for a while: in some cases, vendors' "UTF" managed services offers are larger strategic threats to telco business models than so-called "OTT" providers. Loss of control over the network might mean Opex savings, but it introduces massive opportunity costs from lower flexibility. Conference presentation here - and watch for an upcoming report I've written for Telco 2.0 on the subject.


  • NFC non-payment applications - interactions, not transactions. More likelihood of success, as it could be driven by a million developers doing cool new things, rather than people wanting a slice of transaction values. But dependent on Apple actually launching NFC as a catalyst, mostly.
  • Mobile VoIP as a telephony replacement - we're getting closer, but we're still not *quite* there yet. Still issues with data network coverage, user interaction, numbering, ringing and so forth. Maybe we'll get the real viral trigger in 2012. Hear the issues in more depth at a Future of Voice workshop (and sign up for the newsletter!)
  • Operator reported "voice" revenues - driven more by price bundling & accounting than VoIP at this stage. Some of the recent discussions I've had suggest that operator execs think they're inevitably losing the voice revenue battle. Add in some dry-but-important accounting changes and they're probably right.
  • Development of simultaneous dual-radio LTE + GSM/UMTS phones - In my view, a sensible alternative to CSFB or VoLTE for LTE voice, but it might be a good idea too late (a bit like VoLGA). See this post & comment stream. One sneaking suspicion - the first such device might be a TD-LTE / GSM phone as I suspect that we're a long way behind implementing both CSFB and VoLTE on TDD networks.
  • Android tablets - can't see the point really, unless you get something much cheaper and more interesting like the Kindle Fire.
  • Blackberry resurgence/demise - 2011 as been an awful year for RIM. A big outage. Storm over encryption. The PlayBook is a cool device lacking an essential capability (email). It hasn't capitalised on its BBM franchise well enough either and is being "outcooled" in its core teenage market, while being sidelined by BYOD policies in its legacy enterprise customers. Can it survive 2012 as an independent? I'm a lot more pessimistic now than a year ago. Clearly some organisational issues to deal with, but at least it's actually still making some profit.
  • Possible network overcapacity as "data explosion" fizzles out - Let's watch the stats over the next 6 months. I've a feeling that the Cisco VNI numbers have created a self-denying prophesy - the industry has developed 20+ solutions to the "data explosion" problem, and it seems that at least 10 have worked - perhaps too well.
  • HTML5 impact on mobile networks if (& that's a big if) web apps take over from downloaded apps. Nobody seems to have thought about signalling, overall traffic load etc. I spoke about this at a recent Telco 2.0 event in Singapore, and I'm writing a research note for them for early 2012.
I hope you've found this interesting & thought-provoking. As always, I'd love to hear your opinions or disagreements on any of this. 
I'd also like to suggest that you challenge your own company - I do numerous workshops and advisory projects for operators, investors and vendors on "Disruption". Some of this focuses on individual issues mentioned here (eg "What's the negative story around RCSe?"), while sometimes I'm asked for an across-the-board wake-up call to provoke thought and discussion. Please get in touch via information AT disruptive-analysis DOT com.

Wednesday, December 14, 2011

New! Disruptive Analysis Premium Twitter subscription stream @DApremium - sign up today at an introductory promotional rate

What am I doing? 

I am launching the first analyst paid channel for Twitter: a subscription for "premium" exclusive tweets of my analysis, opinion and market observations.

Why am I doing this?

I've been using Twitter for over a year and a half, posting as @disruptivedean. I was a hold-out for a long time before grudgingly signing up. Although I've got a fairly good "reach" from it, I've been vocal in criticising it. I'd much prefer to just use something more useful like LinkedIn or Quora for "status" updates and more detailed discussions instead.

But unfortunately, Twitter is like tax - as an analyst, you have to grit your teeth and do it, painful, time-consuming and distasteful as it is. I end up spending time on Twitter that could be more profitably spent writing posts on this blog, advising clients or taking briefings. It adds cost, but brings little in the way of value or revenue.

There's a temporary warm feeling of "being part of something", and some undeniable self-validation in counting followers. But that's emotion, not business (and I don't use Twitter for my personal life).

One of the problems is that the main beneficiaries of Twitter are the Fourth Estate - journalists, and their digital equivalents and entourage: bloggers [no, I'm not one], PRs, marketing folk and analyst relations people. It does a lot of their job for them. Unsurprisingly, this group of users then promotes it as being the best thing since sliced bread, since for them, it probably is. As a result, it's become table-stakes, and because I've had to do it, I've tried to do it well.

I spoke recently at an IIAR (Institute of Industry Analyst Relations) debate about social media, and described Twitter's realtime platform as "seductive but irrelevant" to the work of an analyst, with little or no real value. 
So today, I'm hoping to change that. To use a telecoms metaphor, I'm "monetising the opinion pipe". I'm launching a premium Twitter channel, open only to paying subscribers or my existing clients. I'm keeping @disruptivedean but adding @DApremium

How will it work?

Ideally, Twitter (or, preferably, one of its better rivals) would have had a ready-made freemium + revshare platform, but at present I'm having to work it out with a mix of "protected" Tweets and offline payment through other channels. Looking about online, it looks like there were a couple of experiments doing paid Twitter streams around 2009, but I'm not aware of anything that currently does what I want.

This is experimental. It's genuinely disruptive. I'm not aware of anyone else doing this successfully. It's possible that Twitter may take a dim view and try & shut it down. Or it's possible that the model becomes more widely successful and I get bragging rights in a couple of years' time, albeit wishing I'd set it up as a software platform instead & sold it for a huge sum.

Clearly, most analyst firms offer various subscription models for their content and advice. This is not something that Disruptive Analysis has not done to date, preferring to use a mix of free content via this blog, as well as standalone published reports, papers and advisory services. A paid subscription Twitter stream, at a low price, fits into the trend towards providing analyst material in short, pithy, concise bursts.

I will maintain this blog at its current frequency, for longer & more analytical pieces. Depending on the response to @DApremium, I may add extra pay-walled material for subscribers in 2012. I'll also keep my free Twitter stream, especially for discussions and interactions, and as a marketing tool.

What's the deal?

So... what does @DApremium offer?
  • My best insights and discoveries, delivered immediately. My existing blog & Twitter readers will recognise the regular "I told you so" moments. Future epiphanies will be on @DApremium
  • 1000+ exclusive tweets per year (not retweetable, no forwarding allowed)
  • A bias towards tweets which name & analyse specific vendors / organisations / operators
  • The bulk of any conference coverage on @DApremium
  • Answers to specific subscriber questions on @DApremium where feasible
  • Additional content & benefits for @DApremium subscribers
  • The ability to interact with a narrower group of high-level industry figures, with a much better "signal to noise" ratio to the open Twitter. In future, may set up a separate forum on LinkedIn or elsewhere for more full debate & longer-format discussions.
(Incidentally, I'm willing to be quoted about @DApremium, as an example of business model innovation in social media, and monetising supposedly free services)

What are some examples?

I only set up @DApremium on 13th December, and I am putting up this page a day later. The first 20 tweets have been (from first to last):

1. The premium Twitter stream of @disruptivedean, exclusively for subscribers & clients. Vendor analysis, event streams, unique insights etc

2. RT @disruptivedean BBC iPlayer 3G streaming approach points to death of non-consensual video optimisation model bbc.in/uufHaA

3. My dislike of Twitter is well known, but if DApremium enables monetisation of the "opinion pipe", I'll admit to seeing the value in it.

4. 2 telcos clearly embracing Telco-OTT concepts most strategically today: Telefonica & Telenor. Most others have smaller tactical activities

5. EverythingEverywhere had 326TB data traffic in last year bit.ly/uTPsFq but 3UK claims 137TB per DAY bit.ly/vcsBKn

6. I know that 3UK has lot of USB dongles & is still unlimited, but find it hard to believe 50x or 100x data traffic of EE. Miscalculation?

7. BBC using HLS adaptive bitrate streaming: self-optimising content. Will be more important than in-network optimisers see.sc/GHBBrI

8. Ah, seems EverythingEverywhere 326TB traffic just applies to the sharing/roaming deal T-Mo/Orange, not "native" data on each network

9. Still a disparity though: 3UK=137TB with 6m subs, EE=maybe 50TB/day with 27m subs. Big difference is down to market share of 3G USB modems

10. Telcos likely to be net losers on API balance-of-trade with Google, Facebook etc. eg GMaps charging coverage checkers see.sc/JtEH3b

11. Trying to work out whether there are actually any real-world implementations LIPA or SIPTO offload standards yet. Some vendor support.

12. Interesting presentation on RCS-e: suggests Spain launch now early 2012, not 2011 after all. Embarassing for GSMA see.sc/c97hGw

13. Congrats 3GPP for saying user WIFi prefs take priority over MNO steering. From OPIIS docs "Solution shall allow UE to override rules"

14. Thinking a lot about impact of HTML5 on networks. Totally kills app-based policy - there could be 200m different versions of Facebook Mobile

15. 2012 is likely to see dampening of NFC hype. Where it happens, it will be driven by non-financial interactions, not transactions or purchase

16. Wondering if the attempt by network depts & 3GPP to "own" online charging infrastr is the most damaging strategic mistake by mobile industry

17. Nokia Lumia devices seem well-received & even selling out in some places. But is that just because nobody had the confidence to order many?

18. Rogers in Canada extending its subs' numbers & voice services over the web via an OTT-style CounterPath platform t.co/igvIaiz

19. Ericsson's slow & grudging embrace of small cells seems to reflect fears over cannibalisation & commoditisation http://see.sc/sAsqoE

20. Uploaded my intro preso on LTE ("A contrarian's view") that I gave at last week's Layer123 LTE/EPC event in London www.scribd.com/doc/75687759

I can send across some more recent tweets via email if you get in contact with me

What does it cost?

Pricing for a new service like this is difficult. There are few benchmarks. Obviously, full analyst subscriptions can cost $10s of thousands per year, but with much more content. Disruptive Analysis' own research reports (such as January's forthcoming Telco-OTT report) vary from $hundreds to $thousands. At the other end of the spectrum, consumers getting a daily horoscope via SMS can cost $140 a year.

Now, I'd like to think that my insight is worth rather more than an astrologist's, even in volume terms - you can expect at least 3x the number of messages per year.

How can I sign up / pay?

Given the relatively small transaction size, I'm hoping that the bulk of subscribers will use credit cards or Paypal.

EDIT 1st Feb 2012: The introductory discount pricing period has now finished. 
Please go to the new price-plan page here

If for some reason this doesn't work (sometimes PayPal can be a bit flaky), contact me via information AT disruptive-analysis DOT com and I can issue an invoice or payment email request.

You may want to send me an email anyway, or add me on LinkedIn so I can get to know my community a bit better. If you're an existing Disruptive Analysis client, please let me know and I can give you a complimentary subscription.

If you sign up - thanks for valuing my input, and recognising the value in a new form of analyst interaction & advisory delivery model. I'll share feedback as we go along, and I hope you find the service useful & thought-provoking.

The fine print

A full set of terms & conditions are available on request. Key terms include:

- @DApremium Tweets may not be retweeted, copied/pasted into other Twitter streams, forwarded via email or otherwise distributed.
- Subscriptions are per-user, with the expectation that a maximum of 3 people have access to any individual Twitter account. Corporations [eg company-wide accounts @XYZcorp] will need to contact Disruptive Analysis for access
- Unauthorised distribution will result in removal of the subscriber's ability to access @DApremium
- @DApremium subscriptions will last for 12 months from the date of payment. Renewal reminders will be sent 1 month and 1 week before expiry.
- Subscribers recognise the limitations of Twitter's 140-character posts - this means that abbreviations, approximations & concise statements are typical. Complex ideas, trademarks etc may need to be condensed, and nuances may be lost. Any issues with accuracy, clarity or disputes should be raised with Disruptive Analysis, which will use either Twitter, blog or other channels to make clarifications or corrections if necessary.
- Subscribers recognise that paid Twitter streams represent an innovative business model and may be subject to change because of circumstance beyond Disruptive Analysis' control
- In particular, Twitter may change its terms of service, block or delete the @DApremium account, or require a new payment mechanism
- If there are problems or interruptions to the @DApremium service, Disruptive Analysis will attempt to find alternative channels for delivering status updates. These could include LinkedIn, private blogs, other social networks or, if necessary, email lists
- I'll be travelling over the Xmas and New Year period until around Jan 10th, so posts over the next few weeks will be less frequent than during normal periods. This is part of the rationale behind making the intro subscriptions valid until 31st Jan 2013

Thursday, December 08, 2011

Telco-OTT Strategies & Case Studies: Pre-publication discount on forthcoming Disruptive Analysis report


I've discussed for several years (via private advisory work and this blog) the prospect of operators running their own "access independent" services. This means mimicking the major Internet players' role by offering what I'm calling "Telco-OTT" services. In other words, decoupling the network from the application - something which is highly controversial, especially for telecom traditionalists.

Google, Facebook, Skype and others have shown the power and value of standalone applications and services. Finally, operators are exploiting the scale of the multi-billion user Internet, the flexibilty of PCs, and the ease and virality of the smartphone app paradigm, to go beyond the narrow confines of their own access customer base. In other words, they are embracing the doctrine of "If you can't beat 'em, join 'em". But while compelling, it is not easy - there will be failures as well as successes.

I'm now in the final stages of putting together a landmark Disruptive Analysis strategy report on Telco-OTT services. It will span both fixed and mobile operators, and a broad set of applications and case-studies. It will be published in January, but I'm giving advance notice here, a taster of some of the thinking in the report, and offering this blog's readers a promotional discount for pre-orders.

This will be the first analyst report to focus on, describe and identify the trend - and "give it a name".

There are two broad strategies I see emerging for operators' own inhouse OTT activities:
  • Add value to existing access subscribers, by adding service "accessibility" via 3rd-party networks and devices
  • Extend brand and service/content reach to non-access subscribers
A prime example in the first category is fixed-broadband IPTV or cable providers, which are increasingly offering their subscribers a mobile app to watch content while "on the go" - typically on smartphones or tablets connected via other operators' networks. In the communications arena, Telefonica is trialling its O2 Connect VoIP service (based on Jajah) with an initial focus on existing O2 subscribers connecting via WiFi. Even RCSe believers see value in a "broadband access" option for OTT linkage into a user's service, via a PC client and generic Internet connection.

The second class is a more "full" vision of OTT-style services by operators, reaching new users who do not have a "subscription" at all. Orange's ON Voicefeed (a third-party visual voicemail) is a good example, as is T-Mobile US Bobsled (a Facebook-based communications app). Many operators also have traditional web portal or video/audio businesses, selling or delivering content to the Internet at large.

The first strategy is generally lower-risk and easier to monetise; the second is much more difficult (essentially on a par with any other Internet startup) but ultimately could generate more value. The report considers both paths, as well as the different options for organic development and acquisition. It also looks at strategic pitfalls and mistakes for Telco-OTT services, such as the creation of custom hardware (eg Vodafone 360 handsets).

There are also various other offerings that fall into the Telco-OTT envelope - classical enterprise VPN remote access services, femtocells connected over another operator's broadband lines, even WiFi "onload" used as a way for one operator to gain a service "pipe" onto another operator's smartphone fleet. New markets as diverse as M2M and social networking also yield opportunities and examples.


Wednesday, December 07, 2011

What is the cost of VoLTE? Is it a huge strategic error for the mobile industry?

I'm sitting at the Layer123 conference on EPC and LTE in London.

I've just heard a joint presentation on VoLTE from the GSMA and the MSF. The latter was about the recent interop trial, and the former was a general flag-waving / propaganda pitch. Apparently "everyone" is now agreed on VoLTE deployment. (I assume that comes from the same GSMA dictionary as "ubiquitous" used in discussions on RCS / RCSe). The subsequent presentation is from Germany consultancy Detecon, and the former one was from BT Wholesale.

Some outputs so far on LTE Voice:

- General agreement that CSFB still reduces QoE by increasing call setup time and dropping the LTE data connection
- "Pure" VoLTE should be available without handoff when the user moves out of LTE coverage in late 2012 / early 2013, after trials in early/mid 2012
- E911 support is a bit of a hassle for the US and a short-term cludge will be needed
- VoLTE with SRVCC (for handoff to 2G/3G at the edge of coverage) is late and complex
- Massmarket VoLTE with SRVCC "should" be available in 2014. I think that's both (a) delayed from initial discussions, and (b) very optimistic. I think 2016 is more likely when you take into account various other factors like devices

But the elephant in the room is the growing realisation in the industry that mobile voice telephony revenues are declining. A survey of strategists at the London Telco 2.0 event I attended a few weeks ago suggested 20-30% revenue erosion for operators on telephony in the next 3 years. The Asian equivalent in Singapore last week was even more bearish. Private conversations I've had with telco voice and strategy executives also demonstrate pessimism - sometime to the extent that they see telephony revenues evaporating almost entirely, under the onslaught from MVNOs, Google, Skype and the regulators.

So let me get this straight:

- Telephony revenues are declining because of external (and unstoppable) factors
- CSFB gives a worse user experience than circuit telephony
- VoLTE is only useful where there's actually LTE coverage
- Without SRVCC, VoLTE will drop calls (worse QoE than circuit)

Now in theory, VoIP means more calls can be squeezed into a Hz of frequency - so there are efficiencies on the radio and on backhaul capacity. But all the complexity of policy management, QoS, legacy interworking needs to be factored in.

Can anyone honestly say they believe that VoLTE will offer a lower "production cost" for telephony than GSM? Or are the costs per unit about to go up, just when the revenues are going down? And CSFB probably means higher costs AND lower quality than today's GSM.

That doesn't make economic sense - what operator will want to invest in more-expensive technology in order to provide a worse service in a declining market?

EDIT - a couple of people have pointed out that CS voice infrastructure isn't that cheap either, and neither are the legacy billing/OSS systems. True, but modern developing world 2G/3G networks show that CS can be really cheap if deployed today. I think a CS refresh (or full OTT-style service with a chance of Future of Voice type new revenue streams) may be a better bet than VoLTE.