Telcos: time to stop reporting "subscribers" and start counting "users"

Perhaps the singlest greatest block to future growth and success for telecoms operators is their collective insistence on thinking of every customer as a "subscriber". The technology base, the unstated assumptions about business models, the financial reporting, the internal KPIs and the marketing is almost entirely geared towards "subs".

Ironically, even the well-known term ARPU is actually really ARPS.

Whilst it is absolutely true that subscription-type contracts and prepaid accounts will continue to be important, it is also essential that operators wake up to other ways of engaging.

Facebook and Google do not have "subscribers" - they have "users", which may well avail themselves of a variety of services. Magazines have "readers", some of whom are subscribers, others of whom buy a copy on an ad-hoc basis.

Train companies have "passengers" who make a number of "journeys" - some of these are season-ticket holders (ie subscribers), others purchase tickets when they need them, others have their ticket prices deducted from a prepaid credit for travel.

It amazes me that I'm seeing forecasts for LTE "subscribers" for 2015, as though that is the only way customers might purchase service in five years' time. If I buy a high-end camera with an LTE module in 2015, and use it twice a year when I go on holiday, in no way am I a "subscriber" - especially as I may want to lend it to a friend at another point. If I'm in a conference room where the organiser has provided free in-room LTE for a day, are we all subscribers? Including the projector and printer?

Some operators have started to split out M2M connections (mostly because the ARPU is much lower and would drag the averages down). Well done Verizon, for example. But that is only part of the story, as generally even those connections are regularly monthly/annual deals.

With the rise of operators' own social network and Internet properties (Orange ON, Telefonica Jajah, Vodafone 360 etc), they will be obtaining a significant user base that do not buy access. Are they subscribers? If they *do* buy access, are they double-counted as both subscribers and users?

Add to this the problem of two-sided models, where someone else is paying for a service such as a calling or messaging API, and the definitions get even fuzzier.

The bottom line is this: if all you've got is a hammer marked "SIM" then you can only view all your customers as nails. You need a better toolbox.

Mobile traffic management - the Inter-technology war begins

I've been following the proliferation of mobile broadband traffic management technologies for some considerable time now, having published a reseach paper on the bewildering range of options a few months back.

The continued hoo-ha around Net Neutrality is starting to catalyse open warfare between the advocates of different approaches to solving the mobile capacity crunch.

WiFi vs. femtocell offload competition is nothing new, but it's interesting to see new rivalry springing up between other purveyors of policy management and control. Flash Networks, for example, is one of a number of vendors that compress/optimise traffic, especially video destined for transmission over cellular networks.

There is quite a lot that can be done at the interface between the operator's gateway and its main Internet peering point - reformatting video, changing codecs or frame-rates, buffering cleverly and so on. Flash has made a very pointed Tweet to say that this is a much better option than performing brute-force traffic-shaping somewhere else in the operator's core or access network. Basically, they are suggesting it's better to compress traffic actively by modifying, rather than force the "natural" content to squeeze through a narrowed pipe.

(The Flash Tweet in full: Nice illustration of watching video with not enough bandwidth, or why bandwidth SHAPING is a DISASTER for online videos http://ow.ly/2upYv)

Of course, the content-optimisation approach to traffic management is not without its flaws either. Firstly, it changes the data transmitted, which is very much the spirit of Internet connectivity. Secondly, it is typically not "bearer-aware", only compressing the traffic if there is actual congestion. Most such solutions are also blind to whether the device is connected via femto or macro networks as well - there is almost no justification for changing content delivered over the user's home broadband and femto.

I don't want to single out Flash here, it's just an example of what I see as an emerging trend for warfare between the various options for traffic management.

Frankly, until we get closer to what I term "holistic traffic management", none of the options will be without problems. There needs to be much more awareness of content, user, network, device, application, actual congestion, pricing/tiering and numerous other criteria in order to make rational and customer-friendly policy decisions.

If you're interested in working through the mobile broadband policy/traffic minefield, please get in touch.

Is mobile voice being over-valued?

Yesterday I read this Wired article on the "death of the phone call" - essentially pointing out that voice telephony is becoming increasingly sidelined in favour of other modes of communication, many of them asynchronous (SMS, Facebook etc).

This sort of flies in the face of my historical stance that voice is the forgotten saviour of the telecoms industry - overlooked, underestimated and shamefully sidelined in terms of product development. I've agreed with others - my associates at Telco 2.0, visionaries like Martin Geddes and all my friends in the VoIP community - that, surely, there must be a pot of gold left in telephony if only we can de-construct it and make it fit better with personal behaviour or companies' business processes.

Surely, there is value beyond the ever-falling per minute price? Clever voicemail or push voice messaging, HD voice, all manner of web- and cloud-based voice mashups.

I'm beginning to have doubts about this. What if the simple fact is that despite all that innovation, we're collectively coming to realise we don't like talking on the phone that much after all? Especially when we've got an ever-expanding array of tools that enable specific interactions to proceed much more smoothly and with less stress.

That doesn't mean we won't still want to have gossips with friends, chats with family or conference calls with clients. But for other things, especially short "transactions" ("Running 5 mins late, see you at the pub, mine's a lager"), we increasingly prefer modes that have less emotional investment or concentration, even if it takes longer to compose an SMS.

I think this may also mean we over-estimate the inherent value of voice when we calculate telecoms market statistics. It is common to see mobile revenues reported by carriers as split something like:

- Voice
- SMS/MMS
- Non-messaging data

But actually, the amount that gets classified as "voice" also includes all (I think - anyone have a reference to the accounting standards?) of the payback of handset subsidy, and also implicitly the "line rental" which is also used for the SMS and data. That's the fee for occupying a number, a "slot" on the HLR, the cost of administering my billing record and so on.

If we start to value voice less, and SMS and Internet/app access more, shouldn't we re-classify the numbers to reflect this? I'm about at the point where I'd rather have just SMS rather than voice, if I could only pick one.

So, maybe my £45 a month contract ought to be split down as:

£10 - "access fee"
£15 - subsidy repayment
£10 - data [I use maybe 600MB / month]
£6 - SMS [perhaps 300 or so / month]
£4 - voice calls [250 outbound minutes]

Whereas at the moment, the accounts would likely show that as £29 voice, £6 messaging and £10 for data.

Another way might be to re-assign the subsidy and access fee elements in proportion with the other three components, so that the £25 is split in proportion 10/6/4 and reassigned, which would give:

£22.50 data [edit: I'd roughly assign this as £8 email, £5 Facebook, £3 Google search & £4 general web browsing and £2.50 other apps' data use. £zero for downloads & app purchases]
£13.50 SMS
£9.00 voice

I'm not really close enough to the accountancy & auditing side of mobile to understand the nuances of reporting, but the exercise (and the Wired article) has got me wondering if all our assertions about "80% of value is still voice" aren't entirely representative of reality. And if that's the case, it makes it even more ridiculous if we are engineering all future networks with the old "voice comes first" mentality.

Edit - I should point out that most of my longer voice calls & conference calls are done from my landline or PC via Skype/SkypeOut. Also, as I don't work for a large company I don't have "internal" calls with colleagues.

Thinking through this a bit more, the argument I have is that it is *assumed* that people are "voice-primary" in allocating revenues that should really attributable to a broader blend of voice/messaging/data. I'm pretty certain that a good chunk of post-paid mobile contract fees are wrongly considered as voice on this basis, given the growing proportion of people who are "data primary" or "SMS primary". For prepaid, it's generally much easier to allocate revenues as they tend to be per-use (except for the "bundle" prepaid accounts popular with some operators in markets like the US)

The Top 10 Unwarranted Assumptions in Telecoms

Disruptive Analysis' tagline is "don't assume".... so I thought it might be interesting to examine what the main unspoken assumptions are in the telecoms industry as I see it. These are themes and concepts burnt into the mind of the "industry establishment", especially standards bodies and traditional suppliers; they are seen as unwritten laws.

But are they really self-evident and unquestionable? I don't think so. It needs to be remembered that the telecom industry has grown up around the constraints and artificial boundaries of 100-year old technology (numbering, for example, or linking of length of a conversation with value). Many of those unnatural constraints no longer apply in an Internet or IP world - it is possible to more accurately replicate society's interactions - and extend them way beyond normal human modes of communication.

For any telecoms company wanting a continuing role for the next 100 years of the industry, it's worth going back to first principles. It is critical that everything previously taken for granted is reassessed in the light of real human behaviour - because we now have the tools to make communications systems work the way that people do, rather than forcing users to conform to technology's weird limitations.

So, in no particular order, these are the assumptions I think are unwarranted:

1) A “subscription” is the most natural way to engage with, or pay for, communications services

2) The most basic quantum of human communication is “a session”

3) It is entirely rational to expect people to want a single presentation layer or interface, for all their various modes of communication

4) Communications capabilities are best offered as “services” rather than being owned outright, or as features of another product

5) A phonebook is the best metaphor for aggregating all of a person’s contacts and affiliations

6) A phone call – (person A sets up a 2-way voice channel with person B for X minutes) is an accurate representation of human conversation & interaction, and not just a 100-year old best effort

7) People always want to tell the truth (presence, name, context) to others that wish to communicate with them

8) People are genuinely loyal to communications service providers, rather than merely grudgingly tolerant

9) Ubiquity is always more important than exclusivity

10) The quality of a communications function or service is mostly determined by the technical characteristics of the network

I'm sure that there are others - and also plenty of debate to be had around the ones I've listed. I regularly encounter innovators and strategists that *do* question these principles, or actively look to subvert them. But I still see a huge amount of focus put on these undebated tenets, especially among standards- and policy-makers. In my view, such assumption is an abdication of duty.

Together with its associates, Disruptive Analysis addresses all of these issues and more - and the implications for service providers, investors, regulators and product vendors. Please get in touch via information AT disruptive-analysis DOT com if you would like more detail on workshops or advisory engagements.

The hidden secret in the Google / Verizon statement on Net Neutrality

I've worked out the "dirty little secret", hidden in plain sight.

On the fixed Internet, Google doesn't actually care about prioritisation. In almost all developed markets, "best-effort" Internet access is perfectly adequate for YouTube and other "heavy" Google apps, especially when used with variable definition-quality, either at the user's control or using dynamic rate adaptation.

The agreement with Verizon means that VZ won't deliberately block or degrade YouTube or other content. That's one worry off the table for Google. But, interestingly, it also suggests that it can't prioritise either - something I'd been seeing expecting as a possibility for broadband business models. At least, it can't prioritise *Internet* traffic. More on the non-Internet service options later.

But I don't think Google really cares *that much* about getting HD YouTube to peoples' PCs, so the fact it won't be able to buy high-priority, QoS-controlled video clips isn't a problem. It wouldn't have bought them anyway - and I suspect Verizon knew that. (Unlike some of the more naive European operators).

Instead, it cares about two things:

• Better targetting of PC-based YouTube advertising (and ads on other Google PC services)
  
• Opportunities for non-PC (and non-mobile) Google services, for TVs, energy meters, tablets, consumer electronics and so forth.

My belief is that the PC is all about the Internet (capital-I). Google wants PC users to carry on seeing adverts in their browsers around search, in Gmail, and importantly, on YouTube and other properties. The Verizon proposals don't limit this. But what I suspect *may* happen is that Google will do deals with ISPs and Telcos on an affiliate-style basis. If Verizon helps Google sell *more* adverts, at higher prices, to YouTube or other users, then it might get a share of the uplift in advertising revenues.

While the proposals stop Verizon asking for money to *prioritise* YouTube traffic, it doesn't stop them asking for money to help *monetise* YouTube traffic. So Verizon customer data + Google adverts on PCs = more money than just just Google adverts alone. Win-Win. And not so good for the other web companies who don't have slick affiliate / advertising-targetting / rev-share engines.

The other side is around non-PC access. To be honest, there's not really any need to connect your TV or your electricity meter or your toaster to the "real Internet". You're not going to be browsing, or downloading new apps, or really doing any of the cool, ever-changing things the Internet permits. They will also have some very specific network requirements - eg the meter or a healthcare gadget will *absolutely* need security and prioritisation - and in fact, having them running as a service separate from the Internet is much safer and possibly regulatorily-mandated. Your TV, on the other hand, will come with extra user expectations - no video buffering, for a start.

So it seems to me that it is quite right that a future Google Smart Grid, or Google TV-YouTube, or Google HealthMonitor service should be able to operate on a completely separate basis from normal, best-efforts, PC-application Google stuff. With appropriate business models, however configured by all the parties involved. And not just Google, either. I wouldn't want a future Verizon / GE remote-monitored pacemaker service to be viewed "neutrally" on a par with the Internet access partition on the network.

The issue here isn't about a two-tiered Internet.

It's that networked services and applications are already split between Internet and non-Internet, and they're going to stay that way.

The Internet is absolutely fantastic for many applications, and absolutely wrong for others.

In the same way, massmarket cellular networks are fantastic for many applications, but I wouldn't want my smart meter or a policeman's radio running over those, either.

This issue of prioritising non-Internet uses of broadband lines is covered in considerable detail in the report on Broadband Busines Models published a few months ago. Details here

Work in Progress (comments welcome) - Code of Conduct for Policy Management & Net Neutrality

A food analogy: I see no reason why telcos shouldn't be allowed to offer "Internet-flavoured access", or "Processed Internet Substitute" if they wish, perhaps "Now fortified with Extra YouTube!".

But they shouldn't be able to sell or market those services as 100% Certified Prime Internet Access.

Like everyone in the telecoms industry, I encounter the issue of Net Neutrality and policy management very frequently. I've been watching developments recently, with the Google/Verizon announcement, consultations occurring in the UK and Europe, Chile deciding on a "hard" neutral policy, Canada choosing a middle ground and various other endless examples of posturing and lobbying.

It's got me thinking about what might be suitable compromises - a sort of generic "code of conduct" which might be applicable as a set of initial rules and goals, and which shouldn't irritate *too* many people. (Although clearly as we've seen with the Google/Verizon announcement, there's some irreconcilable differences between extreme ends of the spectrum of opinion).

I'm particularly thinking about rules for mobile data, which does have some practical constraints that do not exist in the fixed world. But given my recent work on fixed broadband business models and fibre, I'm also interested in more generic principles across the telecom industry.

The key areas I'm considering are:

• Deliberate prioritisation, for example by guaranteeing QoS or providing an SLA for specific services, perhaps in terms of bandwidth, latency, jitter or other variables
• Deliberate de-prioritisation, for example by blocking, throttling or otherwise interfering with a given data stream or service
• Internet-based and non-Internet services
• Internet access vs. other services
• The problems of shared network resources being used simultaneously for Internet and non-Internet services

In general, it depends if there is congestion and contention for resources – if you prioritise something on an almost-empty network it obviously doesn’t impact anything else. Conversely, whether a telco should ever be allowed to constrain an application on an empty network is much more contentious.

My opinions (at this moment, and subject to debate and change) are:

• Operators purely offering "in-house" managed services can do what they like in terms of managing their networks for non-Internet services - for example, on the core network it is fine to prioritise corporate VPNs, and on access lines they can tune their own IPTV or broadband TV services.
• Where resources are used for both Internet access and non Internet services (eg home broadband lines, cell-site backhaul) and in particular where they are shared between multiple users, things get more complicated. In theory, it is OK with best-efforts "full Internet access" being at the bottom of the pile after other services have been dealt with. But there needs to be clarity in marketing, management and oversight. To use an analogy - if you buy a standby ticket for a flight, you know you might not get on. But if the plane takes off with empty seats and you're still refused boarding, you've got grounds for complaint & redress. Best-effort means that the service provider *really* has to make their "best effort" to accommodate you - *and* is still subject to the original contractual terms and subsequent reporting.
• Internet access should be sold on the basis of average real-world speeds, not theoretical peak rates. Ideally, these averages (or perhaps minimum speeds) should be estimated for specific customers or locations (eg "given your house location, you should expect to get an average 2.7Mbit/s in normal use"). Providers should collect and publish actual average achieved rates, to allow comparison with their claims and promises. The stated averages in force at the start of a user's contract should not be allowed to fall during the course of the contract - this means that any subsequent sales of "managed" services on the same shared resources should be *incremental* to existing promised and contractual Internet capacity (calculated via statistical means)
• The term "Internet access" needs some form of "Appellation contrôlée" - it can only be sold and branded as Internet access if it is not subject to undue policy management control, specifically around blocking/prioritisation based on application, flow, or destination. Anything else needs to be marketed and sold differently, eg "Internet-like service" or "Processed & reconstituted Internet, NOW WITH EXTRA YOUTUBE!". Consumer protection agencies need to be at the top of their game to ensure that end-users understand the difference - lessons from the food & drink industries should help.
• *However*, it is fine for the provider to sell different classes of Internet access service, for example by speed, with caps or usage limits or other similar plan characteristic. An acceptable proxy for achieving application-level policy might be to have device-specific data plans, in the knowledge that (for example) BlackBerry users will typically behave differently to those dongles or tablets.
• If capacity is available on the "Internet access" part of the network, it should be illegal to discriminate between different applications provided within that partition.
• Any managed services (eg prioritised YouTube) should be provided by capacity in addition to that already dedicated to contracted Internet access. (Statistical estimations are fine for network dimensioning, as average speeds will be reported subsequently).
• Renting broadband Internet access (fixed or mobile) should be similar to renting a property. The landlord should allow you "quiet enjoyment" without undue interference or limits, but is permitted to enter in emergencies, specify upfront rules (no pets) and hold you to account for destructive behaviour (withholding a deposit)
• Governments and regulators may decide to mandate that all broadband providers give the option of "proper" Internet access, as well as any other managed services. This applies particularly in locations where it has been determined that Internet access is beneficial for economic or social reasons. It should be permissible to price pure Internet services at a higher level to managed Internet-like services.
• “Negative” traffic management [throttling] on a per-application/per-service, on Internet access, is acceptable if (and only if) there are genuine resource constraints or costs, not just because an operator doesn’t like a given site/service
• “Positive” traffic management is fine for non-Internet services (eg operator-hosted IPTV) provided over the same copper or radio as Internet access, as long as this is transparently described in detail to the customer
• "Positive” traffic management [prioritisation] on a per-app basis, on Internet access is acceptable, as long as there are not unreasonable negative impacts on other apps/traffic. There must be absolute, detailed and realtime information about policy available to the end user and all application providers”
• Negative / positive traffic management based on user, location, tariff, device type is generally OK, as long as it is not application/site specific [although these are often good proxies for applications]
• Traffic management to manage integrity of network (eg DDoS attacks) or for legal requirements (regulatory prohibition of VoIP or services delivered to children) is fine
• There needs to be an open “congestion API” provided so that telcos can be audited for fairness of traffic management policies
• It should be illegal to block, prioritise or otherwise interfere with 3rd-party tools and services for measuring network quality, unless they themselves generate unreasonable traffic levels.
• Operators must confirm that peering / transit partners apply equivalent rules
  

OK, I think that's a decent starting point. I'll hope to refine / iterate / clarify these over time. I suspect they will annoy both operators and hardcore Internet advocates. If I annoy both groups equally, I reckon I will have succeeded....

Device rental as a mechanism for mitigating roaming rip-offs

I had a meeting in a London hotel this morning, and saw a leaflet from this company at the concierge's desk. It was titled "Free Unlimited Internet - Rent an iPhone 3GS", with rates at between £12 and £18 per day, including £5 of outbound calling credit. The devices come pre-loaded with the London Lonely Planet and assorted other useful apps.

It follows on from another article I read recently about MiFi rental (£19 for 3 days), with the device sent to your hotel.

Both of these are welcome examples of a concerted push-back against the still-ridiculous prices for mobile data roaming (and of course the continued premium for roaming voice).

It is ridiculous that in many places, data roaming still fails my "taxi test" - Can you walk from point A to point B across a city, using Google Maps on your phone, with the data costs lower than the price of just jumping in a cab instead?

It still astonishes me that mobile operators can, on one hand, profess to being interested in "customer advocacy" to drive "loyalty", while on the other, they will blatantly charge roaming fees so egregious that they constitute contempt for their own customers. (Or, by proxy, charging ridiculous wholesale rates for inbound users, so their home operators have little choice but to scalp them in turn).

Over time, I expect these type of arbitrage opportunities to proliferate - easier-to-obtain SIMs (perhaps pre-registered by your hotel, as they have your ID and passport details). Cheap Android or other smartphones provided by the tourist agency or local stores. Clever call-forwarding options for voice, and so on.

I've been expecting the operators themselves to start developing more intelligent, transaction-based deals ("One week & 1GB for 15 Euros") to address this latent need, but thus far they've abdicated that marketspace. I expect to see a thriving community of service providers such as the ones I've mentioned to fill the gap - although it wouldn't surprise me if Google or Nokia decided to consolidate the space and offer an international managed-service version of this themselves.

In a way it's ridiculous, but I could easily see myself renting a second iPhone or MiFi when I travel, and switching off the roaming on my normal one. I'm sure I won't be alone.

BlackBerry BBM intercept - workarounds probable?

Quick post / question here.

This week has seen half the world's more paranoid and authoritarian countries make announcements about intercepting BlackBerry traffic. The main concern seems to be a belated recognition that the proprietary BBM IM service currently encrypted end-to-end from device to device - and in any case transits the RIM network and is processed in its data centres. Therefore unlike SMS traffic, it's not observable by the spooks - particularly those in countries without access to vaults-full of supercomputers.

It's conspicuous that RIM's announcements appear to have focused on damage-limitation in terms of any feares of its corporate customer base using BES/email service - whereas my read is that most of these governments are more worried about the new army of young BlackBerry *BIS* (Internet service) customers.

Presumably though, the fact that the data transits RIM's network is actually less of an issue than the fact that it's encrypted on the way.

Now, a question for BlackBerry developers: is there a BBM API on the handset in the new SDK and OS version? Because if so, presumably someone will just create a lightweight 3rd-party app front-end to BBM, which encrypts/decrypts everything locally? At which point it doesn't matter whether RIM routes the traffic to Canada or straight through the operator's core network, as it'll still look like gibberish. Come to think of it, the same app could probably do SMS crypto as well.

I have a feeling that this whole "interception" approach may backfire spectacularly on those governments trying to enforce it. This could just catalyse the whole market for private crypto solutions, not just on BlackBerries, but on all smartphones. Next up, massmarket encrypted voice - it's already available for BlackBerries for corporate usage.

It would be deeply ironic if such an authoritarian move prompted a huge shift which ultimately resulted in more privacy rather than less.

Edit - looks like the Bahrain government has a more realistic view of the situation

Device-specific data plans and policy management

There is currently a clear shift among some operators towards device-specific data plans. In many ways, this is not new – 3G modem data for PCs has often had different price points to handset dataplans, for example. This has enabled operators to pitch 3G dongles against DSL competitors, reflecting typical higher monthly usage and so on. It has also reflected the lack of a voice plan sold concurrently, lack of (or lower) device subscriptions and so on. There have also been dedicated BlackBerry plans for many operators, incorporating BES / BIS email connectivity, or just reflecting the specific deals cut with RIM.

But increasingly, data plans are becoming more granular still – a trend likely to continue as we gain new device form factors. iPhone and iPad plans are specific to those products – and easily enforceable (for now) through the use of MicroSIMs which cannot be swapped around. MiFi products, which are inherently multi-device tethers, may also be subject to different plans.

As an example, take 3UK

- iPad MicroSIM only plans: 1GB @ £7.50 / month or 10GB @ £15 / month [1-month rolling]

- Laptop SIM-only plan : 5GB @ £15 / month

- Handset SIM-only Internet plan : 1GB @ £5 / month (which also includes circuit-based Skype calling)

With new devices using prepay and non-subscription models, we can expect to see even more granularity – a navigation device might come with a year’s free traffic reports including data, while a camera might be bundled with 1000 wireless photo uploads. Add in plans which include or exclude WiFi or femto data, and it gets murkier still.

The interesting thing here is that, in essence, we are getting a sort of blurry policy management and mobile traffic management by the back door. Although the correlations are not perfect, typical iPhone usage is different to typical BlackBerry usage, or assorted other products. Less / more video, less / more social networking , less / more web browsing, more / fewer notifications and so on. It’s quite easy to skew the prices and tiers to favour the less network-hungry products – or implicitly reward manufacturers for creating “non-aggressive” devices that don’t hammer the RNCs with signalling traffic so much.

What’s less clear is whether prioritising *device types* traffic is the same in terms of Net Neutrality as prioritising *application types*. Is it fair, reasonable or legal to distinguish between them? Even if they are not dynamically prioritised, it could be possible to rate-limit them - for example peak speeds of 1Mbit/s download vs. 3MBit/s. Under absolute purist views on Net Neutrality, it would probably also fall foul of the strict rule-making. But as we perhaps move towards some more negotiated, nuanced, intermediate arrangements, this is one particular Devil that should be included in the detail.

It's certainly much easier to distinguish between device types than application types in the network.

There are also some interesting wrinkles about what happens when users SIM-swap. I already do this, putting my dongle SIM into a vanilla phone when roaming as the prices are better, and I’m never going to run my laptop over 3G in a foreign country under any circumstances. There are also interesting issues about what happens when new apps are released that change consumption profile – or a major OS/firmware upgrade. In other words, there’s a policy management and enforcement angle as well.

Definitely an area to watch.