In particular, there’s one graph in there that blew my mind, I’m reproducing it here.
Source: Mobidia / Informa (2013)
This shows the proportion of mobile data traffic that is offloaded to private wifi networks (in blue) or public wifi networks (in light blue). Basically, only a quarter to a third of the data traffic consumed by mobile devices is actually delivered over mobile networks (except in Japan and India where it’s half).
Take that, “we’ll only need mobile networks in the future” posse…
The tech press has been abuzz last week when it was first leaked and later announced that Deutsche Telekom would soon apply data caps to their wireline broadband offers (see this Fierce Telecom article for details.) Unlike AT&T style caps, heavy users will not be charged overage, they will be throttled to service levels marginally higher than what we’d get in the days of dial-up.
The company, as is often the case with these stories, claims that this is to avoid the cost of bandwidth hogs spilling over to the general public’s subscriptions. It’s not.
Caps serve no purpose in managing traffic flows, as Diffraction Analysis clearly demonstrated in our study last year entitled Do Data Caps Punish the Wrong Users. In fact, that’s even been admitted semi-officially by the head of the US Cable lobbying association. Data Caps serve no purpose other than to create a very strong disincentive for customers to consume video “over the top”. The fact that DTs own video-on-demand service will not count as part of the monthly traffic allowance is to be expected, and is a clear giveaway.
DT is playing foul, but that is also to be expected: this is just a new front in the war against Net Neutrality. Since operators, incumbents in particular, can’t get a clear go-ahead on the ability to throttle online service providers to their hearts’ content or to make them pay a toll for delivering traffic to end-users who have already paid for the right to access that content, they’re creating barriers on the side of the end-users to make their own service offerings unfairly competitive.
The real question is what happens next: will DT lobby the government and regulator to apply the same caps to their wholesale bitstream offers ? I suspect they will, just as it happened in Canada. Otherwise, other operators in the market will start advertising no-capping policies, and if they’re smart they’ll even start partnering with online content providers to drive the difference (for more on that see Diffraction Analysis’ latest report Building the Optimal NGA Service Portfolio). That could mean loss of market share for DT.
Is the German market truly competitive ? Guess we’ll know soon enough.
The debate on net neutrality has been heated these last few months, with various initiatives both at policy levels and ISP levels highlighting the need to clarify how internet traffic is carried so that journalists and members of the public can form an opinion on these issues for themselves. A while ago, I was asked by Google to write a paper aiming to do just that. It was released yesterday under the title There’s no economic imperative to reconsider an open internet.
I’m quite proud about this. Working on this paper gave me the opportunity to really explore the net discrimination arguments and examine their worth. My conclusions are expressed in the title: there is no big issue related to the cost of traffic management, even as the traffic itself increases. I hope you find the paper interesting, and feel free to spread it around.
For the last few weeks, I have been busy working on a big in-depth report on NGA services. Although its importance can sometimes be exaggerated, TV content is of course at the heart of NGA portfolio considerations, and I wanted to delve a little bit here into how I have examined that issue.
The notion that broadband providers are scared of on-demand TV is not new, and it’s an understatement. All of the recent assaults that we have seen on Net Neutrality from ISPs around the world can be summarized in two issues: SMS and TV. The latter is of course of more relevance to wireline NGA networks, and this is what I wanted to focus on today.
I think the decision to offer a TV service is not as clear-cut as it’s often made to be, and I think many broadband providers may have gone down that road a little faster than was reasonable. TV delivery will only work if you can get the scale to operate it profitably, and even then the profits are likely to be limited. However, if you decide to deliver TV, then you are, like it or not, competing with the various Over-The-Top options available today. And if you decide to compete, you should make damn well sure your product is better.
My biggest frustration with IPTV strategies is this strange notion that it operates in a vacuum; that because a TV is connected to the broadband provider’s network that customer is somehow captive for TV. That’s absurd: with the multiplication of tablets, PCs and other screens in the home, there are now many alternatives to IPTV. If they provide a better quality of experience, then these will be used instead of the TV. And if that goes on for too long, then the IPTV service itself is at risk.
When you look at the marketing of TV offerings over NGA around the world, you see that the first thing put forward is quantity of content: number of channels, number of VoD movies, etc. That’s all well and good, but it’s meaningless to the customer if the ergonomy of the service doesn’t allow him to watch what he wants to see when he’s available to see it as opposed to watching something that happens to be on. And that’s where OTT offers tend to be vastly superior to broadband providers’.
What it takes to compete with OTT offers, in my opinion, is the following:
rich and well-indexed content
fast and intuitive ergonomy
a powerful search engine
an effective recommendation engine
I haven’t tested many broadband provider TV offers around the world, but none of those I did test had all four of these. In fact, very few of them had any of these.
To highlight the different operating paradigms of broadband providers and OTT TV providers, it’s good to remember that in 2007 Netflix opened its recommendation code to developers in the hope of getting improvements on recommendations, and offered $1m to the first team that would get a 10% improvement on recommendation accuracy. Last year, Netflix announced that 75% of their traffic was driven by the recommendation engine. The mindset here is to never let you go.
I remember when I used to work for mobile operators a few years ago and every screen design was conceived to maximise effectiveness, minimise lag and keep the consumer engaged. I don’t think many broadband provider TV interfaces are built that way, and if they don’t radically change the way they approach things, then they’re not competing.
To be fair, I see a few players starting to really get this. In January I talked briefly with the CEO of NDS, now part of Cisco, a company that offers such sleek and intelligent interfaces for broadband providers who want to deliver a high-engagement TV experience. The following video is an example of what Portuguese operator ZON has done with the platform:
Now I know this is an advert, and I don’t know if the quality of experience is really that good when you’re on the end of the line. And this isn’t an endorsement of NDS either: I don’t have the technical background to assess if their solution is better than others’. I just like what I see here.
What I do know though is that ZON offers this only with their fiber access, and I know why: you need a hell of a low-latency to make things this fluid. And NGA is what it takes to deliver low-latency.
At least ZON seems to be asking themselves the right questions to get customers engaged with the TV product, which will make them a lot less susceptible to go looking for OTT alternatives.
In the service report Diffraction Analysis published last week, Building the Optimal NGA Service Portfolio, we not only explore the field of NGA services of all kinds, we also examine the best delivery approaches for each. TV is only part of the equation!
The internet is disrupting so many established businesses that it’s sometimes hard to pull all of the threads together to get a good grasp of where things are going. I was recently pointed to this recent slide deck published by Business Insider entitled the Future of Digital. It’s not without its flaws (too US-centric, some of the data layouts questionable), but it’s the best attempt at pulling it all together I’ve seen so far. I strongly recommend reading it. I extracted one slide from it last week in my post on delinearization. Here’s another slide that got me pondering…
I was just talking yesterday about two French ISPs having given up on building their own linear content packages. One might argue that they did this in part because they see linear TV on a downward slope. The news last night about the Disney / Netflix deal will probably comfort them in that line of reasoning…
Earlier this year it looked like Netflix was facing an increasingly uphill battle to maintain its rich content at such a low price to consumers, and some of the large majors seemed to be poised to bet on cable as opposed to Netflix. But last night Disney and Netflix announced a multi-year agreement for Netflix to distribute Disney movies in the earliest Pay TV slot on their US platform. Of course, we have no idea what the financial aspects of the deal are, and whether in the long term such deals will drive Netflix’s costs (and therefore their prices) up. Still, it certainly puts a halt to rumours that Netflix would not be able to negociate access to such contents.
And as the slide above shows, 16% of US TV sets were used at prime time for non-linear video viewing 4 years ago, it’s now 33%. Maybe Disney sees the writing on the wall as well.
Meanwhile, over in France, Youtube is set to massively disrupt established television ecosystems with the launch of it’s Ligue 1 channel. Sure, they won’t be able to broadcast live soccer anytime soon, but they will I suspect swiftly undermine all the linear soccer TV shows by not only performing the same commentary and punditry roles that these shows have but also offering users multiple ways of watching exactly what they want when they want. Already you can view the content through teams, goals, fouls and high points, and I suspect soon you might be able to look at individual player performances and commentary and probably many other approaches that – as a soccer ignoramus – I can’t even devise.
All of this has left me wondering if the massive shift to de-linearized content on the TV that I anticipated in about 5 years time isn’t accelerating. Which of course raises issues of available bandwidth (all of that content is over-the-top) both in the access and at interconnexion points…
Late last week, French regulator ARCEP published the latest numbers for French FTTH. They’re not looking particularly sexy in terms of acquisition. Not that this comes as a surprise, considering how little marketing and sales action has been happening. However, this article in PC Impact entitled FTTH: Les Concurrents d’Orange Ne Recrutent Presque Pas d’Abonnés (FTTH: Orange’s Competitors Recruit Virtually No Subscribers) highlights a new trend which was anticipated but not yet seem, namely that only Orange, the incumbent, is actually trying to sell the technology. The summary table put together by PC Impact does not leave room for misinterpretation:
During the earlier parts of the year, Orange’s sales represented about half of the (very low) total sales of FTTH in France, which you would expect of a player whose market share in broadband is roughly 50%. But in the last recorded quarter (Q3), Orange’s sales have shot up while their competitors’ (essentially Free and SFR, Numéricable’s FTTLA offering is tracked separately) have dwindled to nothing.
This isn’t all that surprising for two reasons: unti now, none of the three operators deploying FTTH (or theoretically deploying FTTH) were actively marketing or selling. So each was carried by the momentum of their own brand, and roughly accrued subscribers equal to their broadband market share. In Q3 Orange started to market and actively sell through door to door salespeople. As a consequence, they accrued a larger share than what their brand alone carriers.
However, the overall number of new subscribers haven’t gone up, which suggests that there is either a natural cap due to the addressable market (it has long been known that the very large numbers of homes reported as open for service by ARCEP, over 2 million at the end of Q3, don’t mean these customers can actually be connected but simply that the building they live in has fiber in the basement) or that Orange’s marketing and sales are not really effective to carry over people who are not already interested in getting FTTH from any provider. Unsurprisingly, there’s no Field of Dreams effect.
If you want them to come, you have to entice them…
This also means that Free and SFR are sitting idle, having never really embraced FTTH even though Free’s own announcements was at the origin of the whole dynamic for FTTH deployment. Ironically, Free is now advocating for a balanced deployment of FTTH and VDSL (although they were influential in making VDSL persona non grata in the French regulatory landscape 5 years ago) and in this aligned with Orange.
It’s quite clear that Orange’s competitors have dipped their toes in the infrastructure waters and decided it was too cold for their taste. They will now likely wait for Orange to meet its deployment obligations before marketing services that will use Orange’s infrastructures to reach end-customers, just like they have always done. Meanwhile, both Free and SFR have pulled out of TV content curation in favour of French Pay TV giant Canal Plus (see this for SFR’s announcement today)? Please note that this isn’t exactly comparable to Fastweb’s more radical IPTV announcement a few months back. Fastweb pulled out of TV distribution altogether. Free and SFR are happy to distribute Canal Plus’ offers and a basic linear TV package of their own (with some replay features) but have pulled out of VoD completely and no longer offer repackaged TV offers of their own.
Over the years I’ve been fairly critical of Orange both in the way they captured the NGA market and contributed to its slowing down in France and in the way they took a quasi-systematic anti-net neutral stance, even against their own best interests. Their latest service announcements, however, show a change of tracks, and one that I find interesting and mostly compliant with Net Neutrality. That’s not to say they’re not pushing the envelope, but at least they seem to have embraced the paradigm that traffic can stay on your network or it can go elsewhere, and that much of that depends on your own choices as a service provider.
Le Monde published an article today (in French) entitled L’internet “ouvert” d’Orange ferme les portes (Orange’s “open” internet closes doors) which I think misses the point but at least lists the main evolutions in services offered. In a nutshell, here are the big announcements:
a new super duper set-top box which… catches up with the latest evolutions in Free’s and Numéricable’s own set-top boxes. So not much there.
a million FTTH customers as a target for 2014. The article states this as ambitious considering Orange only has 150 000 customers today, but it’s actually fairly conservative considering there’s no advertising for FTTH of any significance today and the deployment targets by 2014 are at least 5 times that.
a new Orange Cloud service which is effectively a Dropbox by Orange. This is something that I’m surprised is coming so late, and that I think is both smart and commercially viable. It’s a natural extension of your broadband service and – provided the T&Cs are intelligently done and the sync clients work fine – should be successful. It’s not clear to me whether the 50GB offer is free and the 100GB offer not-free or if both are optional, I’ll have to investigate that. Orange mobile customers won’t have that traffic count against their caps (which is net-neutrality compliant since it doesn’t leave their network).
Orange wants to integrate more over the top services within its offerings (and not just Deezer and Dailymotion which they own) and have announced a big partnership with Akamai to maximise the hosting of said services in their network. Again, this is smart and totally net-neutrality compliant. It’s a way to enhance the quality of the service they offer to end-users and minimize the amount of peering/transit required to do so.
All of this, in my opinion, is smart. What goes untold of course is what (if anything) happens to genuine internet traffic ? If something isn’t cached in Orange’s network, will you still access it with the same quality of service as you do today? If the answer is yes, then I applaud Orange and would even consider (as a customer) switching to their network just for that reason. If of course the unsaid flip side of this is that they increase contention on the internet access part of the service, then I’d find this appalling.
Hopefully we’ll know more in the coming days and weeks.
Two scientists from Akamai recently published a paper entitled Video Stream Quality Impacts Viewer Behavior: Inferring Causality Using Quasi-Experimental Designs. Now Akamai being in the business of optimizing internet traffic flows and particularly internet video flows, you’d expect the paper to conclude that low quality in video streams causes users to disconnect, and indeed it does. Still, if you accept the inherent bias that Akamai can only measure videos they’re enabling (and therefore presumably not Youtube or Dailymotion), the methodology seems sound. Here are the main results as quoted from the abstract:
We study the impact of video stream quality on viewer behavior in a scientiﬁc data-driven manner by using extensive traces from Akamai’s streaming network that include 23 million views from 6.7 million unique viewers. We show that viewers start to abandon a video if it takes more than 2 seconds to start up, with each incremental delay of 1 second resulting in a 5.8% increase in the abandonment rate. Further, we show that a moderate amount of interruptions can decrease the average play time of a viewer by a signiﬁcant amount. A viewer who experiences a rebuﬀer delay equal to 1% of the video duration plays 5% less of the video in comparison to a similar viewer who experienced no rebuﬀering. Finally, we show that a viewer who experienced failure is 2.32% less likely to revisit the same site within a week than a similar viewer who did not experience a failure.
Emphasis mine. Now this is nothing new, I think everyone is aware of that on some level. What astounded me was how little tolerance users have for non-delivery of a video. The graph provided in the paper (see above) is quite telling also: it shows not only how fast customers drop off, but the fact that the better connectivity they’re using, the less tolerant they are to latency in getting the content they’re trying to view. Fiber users are the least patient of all.
This leads me to three quick (and hopefully interesting points):
first, as I’ve mentioned before, speed is addictive and users who get used to the comfort of low latency clearly lack patience when content delivery won’t follow,
secondly, as highlighted in Numerama yesterday (quoting the same study, in French), service providers who deliberately throttle video and/or specific video content providers, like Free currently does in France with Youtube are damaging the business of these content players, but also irritating customers. There’s only two ways this can go: either customers don’t care enough to churn and ultimately the Youtubes of this world will have to cave in and fork out to get quality delivery, or the customers care enough and leave visibly enough to force the service provider to actually provision the service properly. I know I’m one of the latter, and the first service provider who will provide me with a good offer from now on I’ll switch to. I’m a heavy Youtube user, and getting decent service on that matters more to me than a reputedly lame blueray player I’ve never used in my set-top box,
finally, I find it very interesting that the tolerance towards mobile is so high. Clearly, there’s no alternative if you’re viewing on a mobile (if you were in a wifi enabled area, you’d be using wifi and it would count as whatever the wireline technology is in the statistics), but I still find it amazing that customers are willing to wait so long for a video on mobile. I’m wondering how long that tolerance will last…
Scales of Justice Brisbane Courts – (CC) Sheba_Also
The coming decisions in December regarding possible changes in Internet regulation by the ITU at the heavy-handed request of European incumbent operator lobbying ETNO has been a growing concern
that I meant to write about, to highlight the so many ways in which their proposals are dangerous to the digital economy ecosystem in general and value-destructive for all players in the ecosystem (including those who are lobbying for it) and the countries who are seduced by the idea of a return to the old telephony models of compensation. I couldn’t find the time to do it in recent weeks, and thanks to Dean Bubley, I won’t have to.
The need to forbid internet access bundling is something I wrote about over a year ago as a clear and understandable solution to solve this once and for all. If customers perceived the clear difference between internet access and managed services, you can bet they would care about net neutrality. To be clear, I’m not advocating the impossibility of a bundle sale, but the need for performance metrics of the internet access part of the service to be explicit.
As an additional piece of reading in line with Dean’s opinions on the laughable AT Kearney report published a couple of years ago and advocating for similar one-sided financial mechanisms (which, as I’ve shown here would make many established online business go bust, even those owned by incumbent operators), you can check out the excellent paper written by Robert Kenny on the topic.