Time to reassess Brew

The mobile operators' increasing obsession with content as the way to maintain or boost margins in the future is pushing content download/delivery platforms out of their niches and into the strategic mainstream. Whether or not we believe that content will really compensate the carriers for the commoditization of the voice market, those with any chance of success need three strengths - good, well branded content partners; appealing handset designs; and a simple and effective software platform for developers and downloaders.

The importance of the third element was spotted at an early stage by Qualcomm, always on the look-out for technologies that will boost uptake of its CDMA mobile network technology.

In mid-2001, it launched its Brew application download product in a bid to establish it as a de facto standard among CDMA operators. It has fallen well short of this goal so far, but as it loses its near-monopoly on the CDMA chip market (Wireless Watch December 4), the company has to pursue growth by penetrating new markets, notably W-CDMA - the 3G version of CDMA rival GSM - and by deriving ever more revenue from its critical intellectual property portfolio. This gives Brew a new importance to the company. It may not generate massive revenues in its own right (these are not broken out, but Qualcomm receives a royalty based on number of downloads), but if it gains popularity with carriers, it can make CDMA more attractive, especially at a time when W-CDMA is still struggling; and it has recently been seen as a Trojan Horse for boosting Qualcomm's presence in markets previously closed to it, notably Europe and GSM.

So, far from putting Brew on the back burner, as many thought Qualcomm should, the company has recently made a string of steps that raise the platform's profile and importance. It has upgraded the software with a particular focus on enterprises, seeing these as a back door into the GSM world; it has made noises about a European push; and it is assembling content and applications partners, most recently a deal this week with AOL.

Comparisons with Java

All these are designed to accelerate Brew's growth and this has become more important than ever before to Qualcomm. The platform is now used by 19 operators, up from eight at the start of the year, and features on about 100 handset models from 21 manufacturers. There are 16m Brew handsets in the world and 63m downloads have been made, claims Qualcomm.

But Brew faces huge competition from its major rival, Java, which is heavily pushed by Sun as an applications and content download platform for carriers and has been the main success story for the Java technology on mobile phones so far. Java's figures deeply overshadow Brew's. While the number of Brew-enabled handsets in the world has doubled this year to 16m so far, they have not reached the 25m predicted by the company at mid-year. Nor are they close to Java's statistics - over 120m Java-enabled handsets have shipped worldwide, up from 50m at the start of the year, according to Java's creator Sun, and there are 200 handset models available from 70 carriers, an increase of 35% in the past six months.

But there are signs that Brew could gain ground. Qualcomm's critical decision came when it said it would integrate Brew and Java more closely, allowing Java applications to run on its platform and supporting Java, and particularly the IBM mobile Java Virtual Machine, on its CDMA chips. Although C++ is the native language for Brew, Qualcomm has made some strides in convincing the Java development community to consider its platform, and many mobile games creators are now using both alternatives according to the target market for a particular program.

The acceptance of Java was undoubtedly an admission that Brew could not defeat Java as a content platform for phones, and would have to accommodate its almost default status among some carriers. However, by removing the either-or choice, Qualcomm also made its technology more attractive and risk-free for operators and gave it the potential to carve out its own niches, even in a Java-centric world.

And Brew does have several advantages over Java. Qualcomm has moved far more rapidly to develop functionality such as application provisioning which makes the operator's life easier. Its long experience of working with carriers in its CDMA business make it far more attuned to the real requirements of this group of companies than Sun.

New strengths

Qualcomm has recently looked like a company always a step ahead of its major rival. The AOL deal is a "significant milestone", it claims, allowing for the distribution of popular AOL services, including Instant Messenger, Mail, ICQ and MapQuest, with Brew handsets, to make these models more appealing to the consumer. The company aims to tie any voguish mobile service in with Brew to make the phones the most feature-rich on the market and so encourage carriers to support the platform. Push to Talk, the popular walkie-talkie style communication method that is being rolled out by various US carriers, is a recent example - as it enjoys a boom, Qualcomm has announced BrewChat, which enables voice over IP and PTT on CDMA2000 handsets. This week, the first handset maker agreed to support this new development - Kyocera, which was the first company ever to launch a Brew handset in 2001.

The cumulative effect of such moves, Qualcomm believes, is to convince carriers that Brew will enhance their efforts to launch attractive data services and so increase average revenue per use and average margin. Its most valuable operator ally is undoubtedly Verizon Wireless, whose CTO Dick Lynch recently said "Brew has leapfrogged Java" and that the platform had been critical to Verizon's aim of rolling out more and more data services. The largest US mobile operator says that users of its Brew-based Get It Now service, which offers 115m ringtones, games, photo messaging and other applications, generate $7.50 per month more, on average, than other subscribers.

This is the kind of statistic that will reawaken operator interest in Brew. The other appeal is that it provides integrated provisioning, billing and application aggregation, making the rapid roll-out of content services far simpler for the carrier. The enterprise or carrier can create its own service on top of this platform and choose from the apps that Qualcomm has aggregated. By contrast, Sun's approach with J2ME is to provide developer tools but it does not - surprisingly - offer a readymade platform. Brew has become increasingly focused on mobile entertainment and on services that involve pay-per-download, where its download and billing platform comes into its own and this is also, of course, the primary focus for the carriers in the consumer market.

Qualcomm says that the advantage of Brew over Java is that Java-based services such as Vodafone Live! require significant inhouse development, while Brew comes as a ready-to-use platform, almost like an outsourced service. US operator Alltel launched its Brew service in under two months, it said. Brew aggregates, distributes and manages content and applications and handles billing and the division of royalties between operator, developer and handset maker.

Europe

Qualcomm even believes it can penetrate the European market on the back of these advantages, which would help in its overall strategy to come in from the cold in the GSMdominated region, with its W-CDMA and hybrid chipsets. In September it appointed its first vice president of European business relations for the Brew unit, Johan Lodenius, who will be based in Sweden. Various other senior executives will be relocated to European countries as Qualcomm aims to build the type of relationships it has with CDMA operators in this unfamiliar turf. "We're trying to put leadership locally in Europe," said Lodenius.

The sign that Europe is ready for Brew is the success of Vodafone Live! believes Lodenius. He said that, until this year, Qualcomm had been focusing on building up Brew in its familiar CDMA world, especially as Europe had been focused on handset styles and brands rather than applications. Now Live! shows the world has shifted and Qualcomm needs to get Brew inside the door while the applications market remains immature, before Java has a chance to become a default - and while carriers need to roll out data services quickly to have any chance of catching up with Live!

Its other route into Europe will be the enterprise.The recently announced version 3.0 of Brew increased the ability to easily tailor applications to different user groups, particularly corporate users and business travellers. Qualcomm believes take-up will be driven by enthusiastic individuals within large companies and that up to five enterprise customers in Europe will be beta testing Brew applications by the end of March 2004.

More important than Europe to Brew is Asia, where CDMA and Qualcomm have greater
presence and where content services are taking off rapidly. China Unicom and Japan's KDDI are key partners, although Brew's initial strength in South Korea suffered a blow when the country's three leading operators, SK Telecom, KTF and LG Telecom, backed by the government, developed WIPI (Wireless Internet Platform for Interoperability) as an alternative to Brew. KTF is estimated to pay $2m-$3m a year in Brew royalties.

In general though, operators' attempts to build their own application delivery systems have been slow to take hold - Sprint being a notable example. The real threat remains Java, whose strongest market is also in Asia. Sun has made some efforts of late to catch up with Brew's strengths, notably with the acquisition of Pixo, a company that makes service provider software to manage, distribute and bill for content downloaded to mobiles using Java, in a similar way to Brew.

It seems time to reassess Brew. Widely written off at the start of this year, it now looks healthy and with significant advantages over its rival. But there is a paradox here. Qualcomm needs Brew more than it has in the past, and is using it as a critical weapon in its aim of boosting CDMA at the same time as moving into new markets such as Europe. But it is doing so in the knowledge that Brew can never be a de facto standard and thus give Qualcomm the kind of influence over handset makers, developers and operators that Symbian gives Nokia.

The presence of Java means that Brew will always be a second platform in a two-horse race. For a software house, that would be an acceptable position, provided that other challengers could be seen off and decent market share retained. For a chipmaker and intellectual property giant, a non-dominant software platform, whatever its technological strengths, must always be a sideline and potentially a damaging distraction.

© Copyright 2003 Wireless Watch

Wireless Watch is published by Rethink Research, a London-based IT publishing and consulting firm. This weekly newsletter delivers in-depth analysis and market research of mobile and wireless for business. Subscription details are here.