While pundits and politicians debate the cost and benefits of the National Broadband Network, a much broader discussion is playing out on the global stage. The worldwide surge in growth of internet usage, particularly for online video, is pushing up data traffic at rates that are outstripping the willingness to invest money into the infrastructure required to carry it.
A solution has yet to present itself, but one thing is clear — somebody is going to have to pay.
AT Kearney principal, Andrea Fumagalli, has been studying the cost imbalances of the internet at the behest of the European telecommunications companies Telefonica, Telecom Italia, France Telecom, and Deutsche Telecom. His mission has been to look at the internet value chain and to think about alternative viable models. The use of the internet is growing rapidly, but the spoils are not being shared equally among the many players involved.
“The current model cannot be sustainable, because you need lots of investment if you want to satisfy the need for traffic from the customer point of view,” Fumagalli says.
“And if you look at the traffic, it is mostly for the delivery of video content, which for the moment is not very well linked to revenues.”
The growth in online traffic is staggering. According to Cisco’s Visual Networking Index results for Australia, annual IP traffic in Australia will grow six-fold at a compound annual growth rate of 41 per cent until 2015, when it will reach 6.2 exabytes per year. In that time internet video traffic will grow 15-fold at a compound annual growth rate of 72 per cent, so that by 2015 it will account for 81 per cent of all consumer internet traffic and equal 15 billion minutes of video content crossing the internet in Australia each month.
Quite simply, the capacity of the internet is not being funded at the same rate as the growth in traffic, as neither the people producing nor consuming the traffic are paying adequately for its carriage. And there is little incentive for the companies providing services over the top of the internet to use the network efficiently.
“On one side you have the end user, where most of the pricing schemes are on a flat base, so there is no real link with traffic and usage,” Fumagalli says. “On the other side, the online service providers do not have such a great incentive to have an efficient use of the network.”
But with traffic growth outstripping investment, clearly somebody will have to pay to upgrade internet capacity. Fumagalli says there are three potential models. Firstly, access service providers could look at increasing prices for consumers. He says an increase in the average revenue per user (ARPU) of 20 per cent could provide the revenue needed. A second option is to extend the concept of quality-ofservice on the internet. “There are some industries that will be interested to have a better service to deliver their content, and probably will be willing to pay a certain amount to have better quality,” Fumagalli says.
This option, however, harkens back to an ongoing debate occurring primarily in the US regarding ‘net neutrality’. Some argue that content companies should be exposed to tiered levels of service quality. Several internet luminaries, including Internet Protocol co-founder, Vint Cerf, and World Wide Web creator, Tim Berners-Lee, have spoken out against such moves.
The third model would see the funding of content delivery networks (CDNs) that bypass much of the internet’s infrastructure to bring content directly to consumers. This model is already in use at CDNs such as Akamai.
Fumagalli doesn’t expect any single model to prevail in the short term, but suggests that it would be wrong for governments to increase regulation to push for a preferred outcome.
Another option is to limit usage. Telecom operators in Australia have long used initiatives to improve the usage of the network such as volumetric pricing, and these are increasingly being considered elsewhere in the world.
Part of the problem for telecommunications companies stems from the massive valuations being applied to companies that offer online services that interface with the user, such as Google, Facebook and Groupon, which are well out of step with their own valuations. Across the board, Fumagalli says, online services providers and the top players in the ‘user-interface’ businesses benefit from a return on capital invested that is more than double that for telecom operators.
For example, as of January 2011 the US-based group buying site Groupon employed an estimated 4500 people and had revenue of $US760 million. By June this year it had achieved a valuation of $US15 billion, and some reports suggest its IPO, when it happens, could come in at as much as $US25 billion.
France Telecom, by comparison, employs about 40 times more people and has 85 times more revenue, but has a market capitalisation of $US54 billion — only three-and-a-half times great than Groupon. Admittedly, the Groupon valuation is speculative, but it shows the stark contrast in investor attitudes between companies that provide services over the internet and those that provide the connectivity these services require.
Not all online companies are getting a free ride, however. Google plans to deploy fibre optic connections at speeds of 1 gigabit per second to up to 500,000 US citizens through its Fiber for Communities initiative. In other parts of the world governments are stepping in, including Australia’s National Broadband Network, which Fumagalli describes as a “big and important initiative”. In New Zealand the government is spending $NZ1.5 billion on its Ultra-Fast Broadband Initiative to bring fast connectivity to 75 per cent of homes over 10 years.
Monetise or die
While content services are receiving massive valuations, the wealth is not being evenly spread amoung content companies. Many media companies — particularly those born before the emergence of the Web — continue to struggle to monetise content online, and companies including News Corporation have implemented pay-walls to directly monetise elements of their content.
Other models for content monetisation are emerging and maturing however, driven by the use of online behavioural advertising technology to increase the yield from online ads.
By placing anonymous tracking cookies in the browsers of visitors, website owners learn more about their visitors and their preferences, and sell this knowledge to advertisers.
It has created a huge opportunity for new data specialists, such as Californian company BlueKai. Its founder and chief executive, Omar Tawakol, says a Web portal selling advertising in an e-mail service might expect to make between 50 cents and $1 for every 1000 ads sold. With the addition of data, however, that price can rise to between $5 and $7 per thousand.
This has raised the ire of privacy advocates, who contend that much of the activity occurs without the user’s consent, prompting calls from within Europe and the US for ‘do-not-track’ registers. Tawakol advocates transparency in the way that data is collected and used, but he says such activity is essential to pay for the content that consumers currently receive free of charge.
“There is one thing they dislike more than ads, and that is paying for the content,” Tawakol says. “If you get rid of data, you are going to have to pay for your content.”
After all, someone has to pay.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.