« Vodafone: A “Total Communications” Company? | Main | The Joy of ‘Functional Separation’ - Panel on Telecom TV »

Broadband Still Sold Like Phone Lines - Time to Change?

We’ve been ploughing through the first 300-odd responses to our survey on the Future of Broadband and there are some very interesting results. (If you’ve not done it yet, just click here — we’ve extended the end date to 30th September as a number of organisations are circulating it internally to stimulate thinking about business models).

I’ve picked on one question that’s given me a bit of a surprise. When a consumer purchases “broadband”, what is it they are actually buying?

Today, on both fixed and mobile networks, the standard means of provisioning is to sell the user an access line. You then get a broadband connection where you can attach any device and use any application. The “line” is identified by a SIM card for mobile networks (at least on GSM-derived networks) or some kind of physical port and/or login credentials on fixed networks.

The chart below shows three possible alternatives to the current status quo, where the act of provisioning is centred on an individual person, device or application.

Summary of combined fixed and mobile respondents for 2012, % of responses.

Taking these alternatives in turn:

  1. Personal access — you authenticate to any device on any network and get access to your services and are responsible for usage. The connectivity follows the user. When at Sprint PCS, this was on the cards for the 1xRTT PCS Vision launch in 2002-3. You would be able to pick up any Sprint 3G phone, log in, and have your usage billed to your standard account — without having to call up customer care to provision additional devices. This proved too complex in the back office, and too difficult a user experience, so never happened. With this kind of provisioning system, the number of devices you can use at once may be limited, and you might be billed accordingly.
  2. Device access — anyone who has the provisioned device in their possession can use it on any network. The ability to use your service with other devices may be limited. You could imagine a future iPhone being like this, once they’re done with exclusivity with individual operators. We’ve got wholesale access bought from all the carriers, so you don’t need to worry about their brands. Just pay us the monthly fee. In some ways, CDMA networks already do this (as there is no separate SIM card — it’s the device that’s provisioned), just CDMA roaming is a minority sport.
  3. Application access — each application (voice client, browser, games) separately authenticates to the network and you pay per activated application. The ability to use other applications is limited. Today an operator might offer an “all you can eat” data plan for mobiles in the knowledge that use of data-hungry applications is low (or outlawed by the terms and conditions). There is a very predictable usage profile for Web and email access, so they price for that. This third form of provisioning would formalise that, making the application itself (not the device) authenticate to the network. This would require some suitable security (like that supplied by most modern processors with a trusted computing module) to prevent fraud, with one application acting as an impostor for another.

The surprise to me was that I thought that application access was rather promising. It enables operators to offer price plans that more closely align to perceived user needs, without carrying so much of the risk of the asterisk after “unlimited*”. After all, there’s only so many web pages or instant messages your thumbs can withstand in a day.

What you are keen on is user-centric connectivity. Sell the user a connected lifestyle, not individual lines. This potentially challenges how data services are priced. Why can’t I use the broadband at my parents’ house at top speed, even if they’ve opted for a low-user slow and capped plan? Why do T-Mobile try to charge me for WIFi access when I’m already a 3G data plan subscriber?

These questions are important because there are whole new classes of device coming on stream, such as small mobile Internet tablets. The industry seems to assume that you’ll just keep on buying more and more line-based data plans as your device arsenal proliferates. Our survey shows considerable enthusiasm for re-thinking how the product is packaged and sold.

You can hear Intel’s VP for Europe present on new business models for new devices at the next Telco 2.0 event next month. And if you still haven’t done the survey, here’s the link again.

Ed - An analysis of the whole survey will be presented during the plenary on 17th October at the Telco 2.0 event.

To share this article easily, please click:


I believe Marc Canter is right. Directory management is the key


"It enables operators to offer price plans that more closely align to perceived user needs, without carrying so much of the risk of the asterisk after "unlimited*"" surprises me as a likely plan in anything like a competitive market, at least on the fixed side.

Since the cost is only slightly related to the application or bandwidth, why wouldn't a competitor simply stick with a high bandwidth plan and take the market. On the fixed side, most operators spend less than 5% of revenues for bandwidth, a fraction of the marketing budget.

Why would customers be willing to go to a plan that charges them much more, unless all operators conspired to refuse a regular plan with plenty of data for whatever use. You have to consistently pull over 100 gaigabits/month before an incumbent DSL or cable guy needs to start starts noticing bandwidth costs.

You, if anyone, appreciate how schemes like this, that work in monopolies, break if competitors aren't completely tame.

p.s. Wireless has higher bandwidth costs, and if anyone has solid numbers for the marginal bandwidth costs on wireless I'd welcome them. On DSL and cable, if you market share supports your own dark fiber network (all incumbents, many unbundlers) marginal bandwidth cost is about 10 cents per gigabyte, dropping.

Application access charging has another disadvantage. It makes pricing non-transparant for users/subscribers.
Where this scheme may be of value or for overshoot customers or in new market contexts. Another opportunity may be in a combined mode, where customers pay for general access (personal, home, device) and for additional, guarnteed, bandwidth per application. IPTV could be an application that could subsidize additional bandwidth (but only for that application.

However combining device access and or application access charges require well defined points of modularity at the access. Structural seperation could prove beneficial to ensure this. But the well defined interface would need to rely on an enabler such as IP.

Post a comment

(To prevent spam, all comments need to be approved by the Telco 2.0 team before appearing. Thanks for waiting.)

Telco 2.0 Strategy Report Out Now: Telco Strategy in the Cloud

Subscribe to this blog

To get blog posts delivered to your inbox, enter your email address:

How we respect your privacy

Subscribe via RSS

Telco 2.0™ Email Newsletter

The free Telco 2.0™ newsletter is published every second week. To subscribe, enter your email address:

Telco 2.0™ is produced by: