James Rosewell writes:

51Degrees.mobi's figures show 10 per cent of web traffic in the UK originates from mobile devices. In India this figure rises to over 90 per cent. With global mobile internet usage expected to grow from 14 million at the end of 2010 to 788 million by the end of 2015, every business needs a strategy for mobile connectivity.

Businesses often consider creating apps for their customers in the hope of driving and capturing customers. These apps often contain a fraction of their products and usually lack the functionality to purchase. The mobile app isn’t adequately fulfilling consumers' needs, and while the business is shouting about how innovative their mobile app is, the rest of their company are secretly disappointed by poor consumer reviews, little return on investment and pricey, on-going maintenance charges.

A desktop application written for Windows 7 can be tested on a HP computer and can be assumed to work equally as well on a Dell, Lenovo, Sony, Samsung or any other vendor’s hardware. The same assumption does not apply to mobile applications. Major differences are experienced across versions or vendor implementations of the same operating system. Testing costs associated with native applications are considerable.

Businesses need to understand the different ways of optimising interaction and ways of communicating with their consumers before investing large sums of money.

There are five major mobile operating systems in use today. Apple's iOS, Android from Google, RIM's Blackberry, Microsoft Windows Phone and Nokia's Symbian. Waiting in the wings are Meego (Intel), WebOS (currently HP) and Bada (Samsung). Each operating system has its own development platform requiring unique source code resulting in very little re-use beyond text, graphics, video and other content. Supporting the five major operating systems will require five unique applications.

Native applications can access the resources of the mobile phone directly, assuming the user has provided permission for them to do so. For example, they can interact with the contact list, send and receive text messages, initiate phone calls or access local storage. They offer a rich user experience.

Unless a business case can be narrowed to support a single operating system, however, native applications will be expensive to build, test, deploy and maintain over the software life-cycle. Even if the operating system can be narrowed it’s likely to change extensively over the next two years. As an example Windows Mobile 6.5 applications will not run on Windows Phone 7.

Device fragmentation within some operating systems, notably Android and to a lesser extent Blackberry and Symbian, present challenges for developers. Design considerations such as screen size and input method all need to be catered for with either separate applications, or ones that automatically adapt to their environment.

Almost every mobile device sold in the past five years comes with a mobile web browser enabling web access. Most businesses will already have a website with an address included in all material from business cards to TV adverts plus supporting activities such as SEO or social media support. A mobile optimised web site built using basic user interface features, sharing the same address and other benefits represents a compelling choice.

Risk aware technology leaders should consider web based mobile solutions building on existing technology architecture and skill sets for consumer focused business cases. Internal business cases could be built for one operating system, but the organisation will be tied to the mobile vendor for the lifetime of the application. Overall, web-based applications are a more cost-efficient adaption for utilising mobile web-based interaction.