As a mobile Internet protocol, i-mode could provide
operators with a means of differentiating their services
in the mobile data market, reckons Kevin Buckley

This year, the 3GSM World Congress in Cannes found the GSM world had well and truly embarked on 3G, with at least one, and usually various operators, having launched services in all the major markets. And not a moment too soon, as voice revenue, everywhere, is under pressure from competitors and, in the case of interconnect rates, from regulators. Data (besides just SMS) is therefore charged with the responsibility of complementing the challenges of declining growth in voice revenues.
In general terms, GSM world operators can be divided into two groups for the purpose of analysing their mobile Internet strategies -- the leaders or, frequently, the top two players in any given market. They will usually have far more subscribers than the rest of the competition, forming a de facto duopoly and vying between themselves for the leadership position, quarter by quarter.
These operators' main challenge is to migrate their huge customer bases smoothly from 2G through 2.5G to 3G. Having learnt from their mistakes with Wireless Access Protocol (WAP) phones, which came to market in 1999-2000 before an ecosystem of well-designed, well-conceived sites existed, they concentrate on building services rather than emphasising the technology. They sell camera phones and music downloads rather than GPRS or UMTS.
As such, they introduce their consumer-oriented mobile Internet offerings as content portals on their GPRS networks, signing up subscribers so that 3G can subsequently be marketed as a speed upgrade.
The other group comprises operators that, for one reason or another, need to differentiate their offering from the rest. Some are new entrants, i.e. groups that have no 2G customer base because they came in at the 3G licensing stage and therefore need to wow potential customers into leaving their existing provider in favour of them. Others already have a 2G business but aspire to become a market-leader and so need to raise their profile as a sexy option for mobile phone users wanting content offerings along with voice.
One way this group can seek to differentiate itself is by promoting the fact that it is offering 3G telephony, running ad campaigns that emphasise the new things that can be done with the more advanced phones in terms of content acquisition, m-commerce transactions or location-based services.
At the same time, the more efficient spectrum utilisation of 3G, compared to earlier generations of technology, means that more voice calls can be put on the same wavelength, a fact the new entrants are exploiting to offer cheap voice services. These are designed to bring customers to their networks, after which it is an easier task to persuade them to start using the mobile Internet function and to acquire content.

i-mode as a differentiator

Another way to stand out from the crowd, and one we are seeing in an increasing number of markets in Europe, is with i-mode. Like WAP, this mobile Internet protocol is an overlay on the network and can operate wherever an IP layer has been deployed, i.e. from 2.5G onwards.
The question all operators now face is: how can they make money from data services? It's all very well for a mobile carrier to say it has diversified beyond voice and into data, but mobile Internet access alone is not enough to bolster revenue. It too is being commoditised as operators start to offer flat rate "all you can eat" services to attract subscribers away from competitors who don't. Remember the cautionary tale of Internet service providers in the wireline world, whose initial promise was blighted as the flat rate, always-on environment grew, forcing them to move to value-added services or die.
Let's begin by separating the provision of such services to business/enterprise customers, who want secure mobile access to key applications running on their corporate networks such as ERP, CRM or SFA, from the marketing of non-voice functions to consumers. The latter represent a mass market that, aside from mobile e-mail and text messaging, essentially boils down to the sale of content. It is the provision of data services to consumers that I want to discuss here.

DoCoMo led the way in Japan

hat business is, of course, in its infancy, but there are interesting lessons from a market we at NEC know very well, namely Japan. NTT DoCoMo is universally acknowledged to have been ahead of its time when it came to content with its development of i-mode, the proprietary technology which, from 2.5G onwards, has successfully built both a large subscriber base (some 44 million in Japan today, or 92 per cent of all DoCoMo subscribers) and a huge pool of vendor sites (about 84,500 right now, of which just under 4,400 are 'official' sites, i.e. ones that pay a 9 per cent commission on sales to DoCoMo, and 4,600 have 3G content). In financial terms, i-mode contributed 25 per cent of DoCoMo's total revenue last year, which is not bad for only its fourth full year in operation.                The model not only works for DoCoMo at home in Japan: it has also licensed the technology to operators in 12 other countries, in one case (KPN in Holland), a carrier in which it holds equity.
Eight of the 12 are in Western Europe: KPN and its subsidiaries E-Plus in Germany and BASE in Belgium, Bouygues in France, Telefonica in Spain, Wind in Italy, mmO2 in the UK, Ireland and Germany and CosmOTE in Greece. O2 plans to deploy i-mode in the UK and Ireland this year and Germany (under a different brand) in 2006.
In other parts of the world, Far EasTone in Taiwan launched in 2002 and Australia's Telstra followed suit last year, while both CellCom in Israel and MTS of Russia have signed up with DoCoMo to launch services.

WAP vs i-mode

Meanwhile other industry heavyweights, such as Vodafone, Orange and T-Mobile, are building services based on WAP gateways, with Vodafone live! the furthest advanced in a number of countries (16 at the end of September '04) and subscribers (11.5 million at that same time). Like i-mode, WAP sits above and is thus independent of the radio access layer, provided the network is IP-enabled (i.e. 2.5G and beyond), they can function to enable a mobile Internet experience.
Before we go any further, let me address the fact that WAP is an open standard while i-mode is proprietary and must therefore be licensed from DoCoMo.
All true, but let us not forget that if, for instance, a games developer wants its game to be playable on the Vodafone live! service, it must write to the operator's proprietary API, called VFX, for the purpose.
In the i-mode world, the main pull for ISVs to write to DoCoMo's API has to date been the carrier's commanding share of the Japanese market. Now that other licensees are coming online there is a buddy group forming which, by virtue of its collective subscriber base, again makes it worthwhile for the software developers and handset manufacturers to work to the i-mode spec. By its size and geographical reach, the group begins to rival the clout of Vodafone.

Street market vs. shopping mall

The difference between the two mobile Internet technologies -- and herein lies the secret of i-mode's success -- is that the latter was developed after the way the market for it would work had been defined, whereas WAP debuted as, to paraphrase Pirandello, a technology in search of a business model.
The i-mode business model can be likened to that of a street market. If a vendor wants to set up a stall (i.e. a site), they agree to pay a percentage of the takings to the local council (i.e. DoCoMo). It is therefore in the operating council's interest to have hundreds of stalls, or indeed thousands, since the Internet does not have the physical restrictions of city streets.
Since the business model was thought out beforehand, from the outset DoCoMo recognised that it was in its interest to promote the take-up of i-mode, and to this end has always allowed so-called unofficial sites to proliferate, i.e. the ones that don't pay the 9 per cent fee on their sales and for whom it does not carry out the billing and revenue collection.
It still makes money on them, however, charging for the traffic generated by their m-commerce activities. © ndeed, some 80,000 of the total 84,500 sites are unofficial, and what they pay to communicate with customers across the DoCoMo network makes up 50 per cent of the carrier's non-voice revenue.
Another major difference between the i-mode model and those of the operators basing their mobile Internet services on WAP gateways is that, in the former, all content acquisition is the result of Internet browsing and all content is delivered via DoCoMo's i-mode platform. In the WAP-based world, there are far less sites and the bulk of content is acquired via SMS. The operators derive revenue from the SMS traffic, of course, but considerably less than they would if their subscribers used the mobile Internet to do their m-commerce, particularly now that SMS and MMS are being bundled into cut-price packages along with voice minutes as competition heats up.
The revenue from the browse-to-buy model comes not, primarily, from the time the subscribers are online, particularly now that many operators are going over to a flat-rate model for Internet access. But by charging the content providers to be the delivery mechanism for their ringtones, weather forecasts, horoscope updates or whatever, and by making it easy for thousands of providers to put up sites.
If i-mode is like a street market then the WAP-based mobile Internet model is like a shopping mall. The number of stores (i.e. sites) is small and the commission the mall owners (i.e. the operators) earn is a lot higher, anywhere from 40 to 60 per cent of vendors' revenue in fact. And since most content is bought by SMS rather than on the Web, one could continue the analogy by saying that most shoppers aren't even entering the mall.
The knock-on effect here is obvious. If I receive an SMS inviting me to buy a snazzy ringtone and I text back to buy one, that's the end of the transaction. If, on the other hand, I get an e-mail with a link to a site where I can download the ringtone, the vendor has far more sell-on or sell-up potential while I'm on the site.
There is also a greater opportunity to create a recurring revenue stream by signing me up to regular service of, say, a new ringtone every week or month, or indeed of multiple ringtones so that I can differentiate between calls from my boss (mental note to self: answer swiftly) and from my mother-in-law (mental note to self: let it go to voicemail). NEC believes that more content brings more users, and more users bring more content.
What's interesting in the Japanese market is that, since i-mode was the first and most successful service there, it has created the country's mobile Internet culture, such that DoCoMo's competitors seek to emulate its business model, even though they're not deploying i-mode.
So while Japanese consumers, thanks to i-mode, browse to buy, their European counterparts text to buy. In the second scenario, the content will often not even go through the operator's servers en route to the subscriber, being stored instead on a content server on the provider's network and delivered by SMS, with the operator deriving revenue only for transporting the text message.

Which strategy will win?

There will probably not be a single winner, as the success of any mobile Internet strategy will not just depend on the technology, or even on the business model alone, but also on the market clout of the operator adopting it and the acumen of the executives running the business environment in which they are operating.
As manufacturers, we at NEC play an important part, as we provide the means to make it all happen! We see a place for both models depending on where a particular operator aspires to be. Our challenge is one of continually being one step ahead -- developing the roadmaps for more cost-effective, high performance platforms and the technology on which either model can fit or indeed, any future models.                       

Kevin Buckley, General Manager NEC (UK),  Mobile Network Solutions Division, can be contacted via e-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it