Which Mobile Phone Platform?
Whenever I start working for people, conversation often turns to how I see the market going. Companies embarking on mobile development like to know what they will be developing will be around in the near future and whether they are backing the correct horse.
The latest news from Gartner is that global phone sales were up 21% in Q2/2005 to 190m units. During the same period, Symbian shipped 7.8 million devices and reading between research data I have from the various places, Windows Mobile Smartphones sold of the order of hundreds of thousands of Smartphones.
Symbian currently only has about 4.1% of the market. Microsoft’s market is negligible. A large proportion of the remaining (proprietary phone OS) market runs JAVA (J2ME). Symbian phones also run J2ME.
Will phones with a proprietary OS (and hence J2ME) continue to dominate? What are Microsoft and Symbian doing to change this?
A main driving factor is the cost to build a phone called the ‘Bill of Materials’. The magic figure is currently about $100 . Any more than this and the phone becomes a low volume ’High end phone’. Symbian phones typically cost $115 to build today (2005).
Consumers are expecting their phones to support more media types (sound, cameras, images, videos etc) and these are becoming harder to add, in a cost-effective manner, to proprietary OS phones. Meanwhile, Symbian has been working with Texas Instruments and Intel to squeeze their OS into a ’1 chip’ solution. That is, have one processor handing both the radio frequency (RF) digital signal processing (DSP) as well as the running applications. Less chips means lower cost to build.
The following graph, from a presentation by Ian Weston shows how Symbian expect to take a greater share of the market (200 million phones) by 2008 by reducing the bill of materials to $78…
Microsoft have taken a similar strategy in that they are aiming for a low bill of materials. Microsoft is pushing a low cost ‘ready-made’ solution called Peabody that phone OEMs can license. However, Microsoft have an additional image problem to crack. Phone OEMs, correctly or incorrectly, associate size, security, legal, licensing and performance problems with the Microsoft image. Hence, unless Microsoft does something radical, for example, splitting off it’s mobile efforts into a new ‘brand’ with new values, it will be an uphill struggle.
It’s important to remember that consumers generally buy a Nokia phone, not a Symbian or Windows Mobile phone. Most don’t care what the underlying OS is. In fact, the majority of the 7.8 million Symbian devices shipped in Q2 2005 were Nokia Series 60 devices. Most of these were upgrades offered by network operators on account of the larger colour screen or megapixel camera.
I think Nokia will continue to be the major seller of Symbian devices and I also think (I don’t actually know!) Symbian’s prediction of 200 million devices by 2008 is based on this. In fact, many Texas Instruments press releases mention Nokia rather than Symbian or other phone OEMs. The majority of the remainder of the market will be proprietary OS devices. Unless Microsoft does something markedly different, Windows Mobile will continue to be suitable for vertical, enterprise applications where integration with existing (Microsoft) systems and significantly easier development may be more important.