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

Friday, May 09, 2014

New Report: Non-Neutral Mobile Broadband Business Models. Pre-publication alert & offer

In the next week, Disruptive Analysis will be publishing a major study into the viability of over 25 forms of "non-neutral" mobile broadband & Internet business models.

Recent weeks have seen huge shifts around the concept of Net Neutrality - the FCC is about to reveal details of new proposals, while the European Parliament voted for a harder version of neutrality than many in the telecoms industry wanted. 

Lobbyists, activists and the wider telecoms and web industries are flooding the media and blog/twitterspheres with commentary, opinion and predictions of what might happen if Net Neutrality is/isn't enforced. Questions, rumours and myths about what may happen, if mysterious and vaguely-defined "Specialised Services" are permitted over broadband networks.

Many observers have confused peering deals (eg Comcast/Netflix) with QoS-prioritised services, even though that is an entirely different issue.

But while the principles get debated, a deeper set of questions are being overlooked. What are the possible outcomes for mobile network operators in a hypothetical world, where discrimination of data traffic on the basis of price or QoS is permitted? What are the realistic use-cases that might get applied? Who might pay for what services or capabilities?

... and the elephant in the room - are any of the non-neutral models actually viable anyway?

Two years ago, Disruptive Analysis published a report giving 10 reasons why the concept of "toll-free apps" was virtually impossible to realise. Today, indeed, none exist. The conclusions were correct.

Now, Disruptive Analysis is broadening its analysis to 6 main categories of "non-neutral" business model, and over 25 sub-types, identifying which has a chance of success, and which will face minor or major challenges.

The main categories considered are:
  • Zero-rated data & applications
  • Sponsored data
  • Application-specific data packages
  • Blocked / impeded applications, "slow lanes"
  • Specialised services, or QoS-based "fast lanes"
  • MVNO & wholesale models
One of the key themes is that new mobile data and Internet models are much harder to create than fixed non-neutral services. The realities of radio networks, device OS's, content/app consumption and policy/core infrastructure are hugely different and much more complex. 

There is also no legacy of "mobile TV & VoD" skewing the landscape, unlike fixed Pay-TV markets with both incumbent/integrated content players/distributors like cable MSOs, and popular "OTT" newcomers like Netflix. 

In addition, fixed broadband usually has no scope for arbitrage for the end-user. Homes only have one connection, unlike the role played by 3rd-party WiFi on wireless devices, which acts to keep the cellular operators "honest" and fair, up to a point.

But despite that, some forms of "non-neutrality" do already work on 3G/4G networks, and others are likely to in the future, where permitted.

While concerns about anti-competitive bundling may need to be considered in future, zero-rating of data for some apps or websites is becoming more prevalent, especially in developing economies. Facebook, Google and others are pitching basic zero-rated data apps, as a way for operators to entice low-end customers to take the first steps towards paying for dataplans. More interestingly, some operators are zero-rating resources such as Wikipedia or other education material - hard for neutrality activists to argue against.

While sponsored data for apps is very challenging, other uses are more feasible, such as the use of paid traffic by advertisers - an idea invented by Disruptive Analysis in 2010. Indeed, while AT&T announced its "sponsored data" concept in January 2014, the only public sign of it so far has been free adverts for a brand of chocolate.

Outside of the strict "Internet data" definition, we will also likely see priority-QoS applied to data used by emergency services on public 4G networks, and perhaps for certain M2M/IoT applications. We may also see differential charging and QoS applied to MVNOs, or integrated wholesale deals in the mould of the Amazon Kindle.

More details on the conclusions and outputs of this study will be released when it is published. But clients who pre-order the report now, will benefit from a free 1-hour conference call about its findings.

The report will be around 85 pages in length, and will include the following chapter sections:

Executive Summary

Mobile Broadband - Wider Trends
  Adoption & saturation
  Traffic growth
  The "OTT threat"
  Lessons from fixed/cable
  Vendor landscape
Neutrality: relevance, regulation & controversy

  Net Neutrality Defined
  House view: Avoiding confirmation bias
  The main arguments for/against Neutrality
  2014: regulatory turmoil & status update
  What is a "specialised service"?
Non-neutral mobile broadband: Generic challenges
  Definition of app/service
  Business model fit
  Device-dependency of apps
  The role of the OS for QoS/non-neutrality
  HTML5 & the web
  The challenge of WiFi
  Telco IT: OSS/BSS issues
  Pricing, selling & marketing
  Network dependencies & standards
  Fixed vs. mobile differences
  Organisational issues
  Unintended consequences
Models of non-neutral mobile broadband
  Analytical framework
  Zero-rated data & applications

   - 4x sub-categories
  Sponsored data
   - 4x sub-categories
  Application-specific data packages
   - 5x sub-categories
  Blocked / impeded applications
   - 5x sub-categories
  Specialised services, or QoS-based "fast lanes"
   - 6x sub-categories
  MVNO & wholesale models
   - 2x sub-categories
  Other/miscellaneous models (eg paid peering)
Conclusions & recommendations
  Can 2-sided business models work for mobile data?
  Which models are the easiest / most profitable / least controversial?
  Recommendations for mobile carriers
  Recommendations for regulators
  Recommendations for network/IT vendors
  Recommendations for investors
  Recommendations for app/content companies

Each of the 26 sub-categories has a table of analysis assessing its viability, based on the criteria shown in the example below. This draws on Disruptive Analysis' unique cross-silo understanding of service provider & application business models, network technology, device & OS architecture, developer/content provider motivation & user behaviour.

The report will be published in the next week. An update (c20 pages) will be issued to purchasers around October/November. Pre-publication advance buyers will received a 1-hour free conference call to discuss the findings, at a mutually-convenient time.

The Non-Neutral Mobile Broadband Business Models report can be ordered immediately via the Buy Now links below, or via information AT disruptive-analysis DOT com - payment accepted by credit card, Paypal and PO/invoice/bank transfer.

For online purchasers, the PDF document will be sent by email, typically within 24hrs of receipt of payment, although sometimes travel schedules may mean a small delay. Please include company name for licence purposes, and VAT number for EU purchasers. 

Non-Neutral MBB report (PDF) 1-3 users

Non-Neutral MBB report (PDF) Corporate Licence


Pal Zarandy said...

Hi Dean,

Does your report cover European examples described below?

(I copied it from our Digital Fuel Monitor website http://dfmonitor.eu/insights/2014_apr_premium_zerorate/)


"Incumbent European telcos are favouring their own or their OTT partners’ messaging, communication, music streaming, video streaming, mobileTV, cloud storage applications by zero-rating the generated volume i.e. volume generated by these applications does not deplete the end-user’s open internet gigabyte volume allowance. Zero-rating is essentially blunt potentially anti-competitive price discrimination. It favours telcos’ own, or their partners', applications and services thereby placing those offered by other competitors at a competitive disadvantage. In markets where big telcos face no challengers, such as Germany, and where the gigabyte prices for open mobile internet access are prohibitively expensive, price discrimination in favour of telcos’ own applications could be a game changer"

Dean Bubley said...


I'm aware of your arguments about zero-rating, but I am not convinced that they are anti-competitive, rather than benign forms of promotion & packaging.

Is there any evidence that zero-rating significantly changes user behaviour in developed markets? For example VF UK inclusive Spotify in some plans, but does not zero-rate it. Does it get lower conversion rates than operators which do?

Similarly,if zero-rating in Germany distorts the market, we should see abnormally strong data consumption growth, despite small/flat caps, as users consume bundled apps in large volumes, rather than restricted open-Internet competitors. Have you seen that reported?

It's something to watch out for, yes, but seeing as a lot of the zero-rated stuff bundled in is rubbish like Joyn, it looks more like a fire-sale than anything to worry about. Quite frankly, I wouldn't expect people to use Joyn even if they were paid to - giving it away for free is more a sign of desperation.


Pal Zarandy said...

Hi Dean,

Are 'fast lanes' (that is differential speed caps) necessarily anti-competitive? Maybe not, if the baseline speed (data rate) cap is high enough that it does not really cause user experience degradation for 'normal' apps.

Are differential volume caps anti-competitive? Maybe not, if the baseline volume allowance is large enough that users can comfortably "consume" their monthly "normal" video streaming, cloud, etc apps without hitting the cap.

Zero-rating very low volume consumption apps (that is setting a special unlimited volume cap to these apps)like instant messaging may be OK. Joyn is increasingly advertised for peer to peer video messaging, so high volume use cases.

To put this Joyn thing into context it is important to note that T-mobile in Germany has as low as 200MB volume cap on its €31.2 smartphone plan. See: http://dfmonitor.eu/DeutscheTelekom/

The point I am making is differential volume caps per app on volume capped Internet access offers (which are typical for mobile internet) are (at least) as problematic as 'fast lanes' that is differential speed caps on speed capped offers (which are typical on fixed-line internet access)


Dean Bubley said...

Fast lanes aren't differential speed caps, they are different packet queues for QoS-prioritised applications or content. Depending how constructed they might just "tilt the playing field" a bit towards certain apps or give them outright priority. Definitely non-neutral - whereas speed *tiering* (ie individuals limited to 2Mbps or whatever) is not.

Joyn might be advertised for all sorts of things, but if it's crap nobody will use it. End of. Doesn't matter if it's free or not. You'll notice that Whatsapp, LINE & co are conspicuously ambivalent about it - it's not a serious rival.

The 200MB cap is not unusual - lots of countries have some low-cap options. Don't forget there's a significant % of people who don't want smartphones or just have sporadic use in mind. It's actually a good thing to have a range of options. And the people who'll want to pick low-end datacaps are precisely the last ones to want advanced apps anyway.

I haven't yet seen evidence that zero-rating telcos' own-brand apps drives significant uptake. If they start producing decent apps then maybe it's a cause for concern.