According to this article on TelecomTV, the Dutch incumbent operator may be about to try to attempt the so-called "personalisation" approach to mobile broadband, charging "per-service" for Internet functions, in an effort to stem the rising use of 3rd-party applications that are substituting for SMS - behind last week's surprise profit warning.
Disruptive Analysis' view is that most app-specific personalisation concepts for mobile data charging are deeply flawed, and will likely lead to churn, counter-measures and outright animosity from otherwise neutral Internet players, as well as probable regulatory intervention. They will also most likely have significant problems with "false positives" and "false negatives". In other words, they will make a bad problem even worse. I wrote a blog post detailing some of the complexities a few months back - here. Although details are still sketchy, it looks like KPN's management may have made a knee-jerk reaction to its poor figures in an attempt to reassure investors.
Although this is in essence a "Net Neutrality" issue, it's not really something where I feel the issue is the principle at stake. It's more that it's a tactic that just won't work - and may quite possibly backfire to the extent it becomes suicidal.
To recap, KPN explained last week that its Dutch consumer wireless revenue shortfall in Q1 2011 was because:
"accelerated changing customer behavior became visible amongst smartphone users. New popular ‘apps’ on smartphones offer alternative ways of communication beyond traditional voice and SMS. The increased usage of these ‘apps’ lead to decreasing SMS and voice usage resulting in lower service revenues."
It further said that "Short term measures are taken to mitigate the impact on service revenues from these trends; these measures include personalized ARPU optimization and reduced discounts on data."
The accompanying presentation to the results shows the icons of Viber, WhatsApp and another service I'm not sure of (on slide #22). I suspect that KPN picked some relatively-minor examples rather than stir the pot by blaming Facebook or BlackBerry Messenger straight away. In other words, what many people have long, long expected has finally started to come true - SMS is starting to be eroded by alternative IP-based messaging firms. I described SMS as a "sitting duck" as long ago as November 2006.
Given that the writing has been on the wall for that long, you might have reasonably imagined that operators and their standards bodies might have thought about creating newer, better versions of messaging to compete. And that doesn't just mean relying on the pipe-dream of IMS RCS or old-style mobile IM - but actually creating something innovative that has the appeal of FB or BBM. Yet KPN only invested €82m in R&D in 2010 in total, against revenues of €13.4bn, much of which probably went on testing LTE and new fibre systems, developing back-office systems and so on. I'd be surprised if more than €10m went into trying to invent new services and applications, either solo or collaboratively. By way of comparison, one of the companies that has KPN scared has only just raised $8m - and so has presumably spent less than $5m so far.
But no.
KPN seems about to try to bill "per-application" for things that compete with SMS. Even leaving aside the inevitable use of email as a work-around, there are dozens of gotchas relating to mashups, VPN tunnels, or even hiding messages steganographically inside images. Will it block PCs from accessing parts of the web as well? What about prepaid data users? Roamers? MVNO subscribers?
I got upgraded to the new Facebook messaging platform the other day, which blends email, IM and FB messages. All inside an SSL tunnel to facebook.com. Packet-inspect that...
Now it's possible that KPN is going to try and do this in a more sensible fashion - perhaps zero-rating "friendly" sites, and increasing data prices for everything else. White-lists are sometimes easier to manage than black-lists.
But I have a sneaking suspicion that KPN has been over-sold on "personalisation" capabilities, and is about to bite off more than it can chew. Becoming "part of the problem" and taxing innovative substitutes will not succeed in competitive markets. Operators need to become "part of the solution" and offer something better. Despite the appealing Dutch folklore, this is not a hole in a leaking dyke than can be plugged by a small boy's DPI-enabled finger.
Disruptive Analysis' view is that most app-specific personalisation concepts for mobile data charging are deeply flawed, and will likely lead to churn, counter-measures and outright animosity from otherwise neutral Internet players, as well as probable regulatory intervention. They will also most likely have significant problems with "false positives" and "false negatives". In other words, they will make a bad problem even worse. I wrote a blog post detailing some of the complexities a few months back - here. Although details are still sketchy, it looks like KPN's management may have made a knee-jerk reaction to its poor figures in an attempt to reassure investors.
Although this is in essence a "Net Neutrality" issue, it's not really something where I feel the issue is the principle at stake. It's more that it's a tactic that just won't work - and may quite possibly backfire to the extent it becomes suicidal.
To recap, KPN explained last week that its Dutch consumer wireless revenue shortfall in Q1 2011 was because:
"accelerated changing customer behavior became visible amongst smartphone users. New popular ‘apps’ on smartphones offer alternative ways of communication beyond traditional voice and SMS. The increased usage of these ‘apps’ lead to decreasing SMS and voice usage resulting in lower service revenues."
It further said that "Short term measures are taken to mitigate the impact on service revenues from these trends; these measures include personalized ARPU optimization and reduced discounts on data."
The accompanying presentation to the results shows the icons of Viber, WhatsApp and another service I'm not sure of (on slide #22). I suspect that KPN picked some relatively-minor examples rather than stir the pot by blaming Facebook or BlackBerry Messenger straight away. In other words, what many people have long, long expected has finally started to come true - SMS is starting to be eroded by alternative IP-based messaging firms. I described SMS as a "sitting duck" as long ago as November 2006.
Given that the writing has been on the wall for that long, you might have reasonably imagined that operators and their standards bodies might have thought about creating newer, better versions of messaging to compete. And that doesn't just mean relying on the pipe-dream of IMS RCS or old-style mobile IM - but actually creating something innovative that has the appeal of FB or BBM. Yet KPN only invested €82m in R&D in 2010 in total, against revenues of €13.4bn, much of which probably went on testing LTE and new fibre systems, developing back-office systems and so on. I'd be surprised if more than €10m went into trying to invent new services and applications, either solo or collaboratively. By way of comparison, one of the companies that has KPN scared has only just raised $8m - and so has presumably spent less than $5m so far.
But no.
KPN seems about to try to bill "per-application" for things that compete with SMS. Even leaving aside the inevitable use of email as a work-around, there are dozens of gotchas relating to mashups, VPN tunnels, or even hiding messages steganographically inside images. Will it block PCs from accessing parts of the web as well? What about prepaid data users? Roamers? MVNO subscribers?
I got upgraded to the new Facebook messaging platform the other day, which blends email, IM and FB messages. All inside an SSL tunnel to facebook.com. Packet-inspect that...
Now it's possible that KPN is going to try and do this in a more sensible fashion - perhaps zero-rating "friendly" sites, and increasing data prices for everything else. White-lists are sometimes easier to manage than black-lists.
But I have a sneaking suspicion that KPN has been over-sold on "personalisation" capabilities, and is about to bite off more than it can chew. Becoming "part of the problem" and taxing innovative substitutes will not succeed in competitive markets. Operators need to become "part of the solution" and offer something better. Despite the appealing Dutch folklore, this is not a hole in a leaking dyke than can be plugged by a small boy's DPI-enabled finger.
I am not surprised. KPN is one of the worst telcos in Europe. I remember when I managed tech ops at an ISP in Amsterdam, they could not provide SS7 interconnect capacity in any reasonable time frame (less than a year) at a majority of their exchanges, and they once cheerfully informed me I would experience 3 months' interruption of service on my $100K/month transatlanctic STM1 circuit.
ReplyDeleteA lot of people attacking operators for trying to defend their positions. It may be true operators need to add value and innovate services, but they are experiencing serious financial problems with mobile broadband.
ReplyDeleteIf regulators eventually decide to avoid such policies, there is also another alternative operators have: stop investing in MBB (if it is not profitable).
What regulators would do then?
Fazal - thanks for your comment
ReplyDeleteDavide - There are plenty of defensive positions. Some will work, some won't.
What's your evidence for "serious financial problems with mobile broadband", anyway? Some (eg 3UK) still seem perfectly happy even with flatrate uncapped pricing.
If you mean "they've mispriced it" then that's a problem of their own creation. In my view, that's certainly been the case for many operators.
They have also lacked the ability to make even basic predictions about possible usage scenarios - I remember presenting at conferences in 2005-6 talking about the future combination of smartphones and HSPA. I remember one senior radio planner telling me that he assumed data growth would stay below 10% per year.
In terms of "what can regulators do?", there's a few options.
The easiest one in my view is to define "Internet access" as a specific named & defined product. An operator should only market & sell "Internet access" if it is neutral & meets specified criteria. It should also be able to sell "Other access" or "Limited access" if it wishes.
The second thing the regulator could do is make it easier / cheaper to deploy networks - eg by encouraging more RAN-sharing, relaxing planning restrictions on new cells, allowing national roaming etc.
The third option is to create a national LTE wholesale network as an overlay - similar to Yota in Russia.
Dean
Davide
ReplyDeleteFor example, it looks like MetroPCS follows the branding concept - offering "LTE access" or "data access" but not explicitly saying "full Internet access" for several tiers of service
http://www.broadbandreports.com/shownews/MetroPCS-Drops-LTE-Intro-Plan-To-40-112063
Fair enough, as long as there's enough competition around. Personally, I'd opt for the full Internet option, probably from a different operator. You might be happy with a limited offer.
There *might* however be a government mandate for Universal Service for at least some level of Internet Access, in much the same way as PSTN today. That will likely be country-specific and also depend on price/availability of fixed broadband Internet access.
Dean,
ReplyDelete>What's your evidence for "serious >financial problems with mobile >broadband", anyway? Some (eg 3UK) >still seem perfectly happy even >with flatrate uncapped pricing.
Well, at least KPN may be an example. Slide 19 of the same slideset you referenced to in your post says "Revenues and other income down 3.1% y-on-y" also because (not only) "lower wireless service revenues from changing customer behaviour".
Some operators have been perhaps too much aggressive in their MBB offers and they will probably reposition their offers by bundling together different services.
I like the idea that regulators can define what is "Internet access", "limited access" and so on. It would make it much easier for consumers to look into the labyrinth of subscriptions offers.
Financial problems?
ReplyDeleteHere in Finland the mobile data consumption per capita was above 700 MByte/month in H2 2010. About ten times higher than in the Netherlands.
There are three "happy pipe" mobile players here with very nice EBITDA margins and CAPEX to sales.
The norm over here is unlimited volume with speed tiers.
The dongle pop penetration is above 34%. MBB connections (excluding smartphones) overtook fixed bb.
Speed (max bitrate) is becoming a better and better proxy for network costs as video traffic prevails. Monthly volume is becoming worse and worse proxy as video makes traffic concentrate more and more on the "prime time". In addition, speed is a much more customer friendly monetization dimension than volume caps, throttling or overages.
Markets like the Netherlands really need some fresh mobile data challengers like Yota in Russia or Three in the UK.