For CIOs & Senior IT Executives

Some General Comments from Meta Group About Mobile Applications

Although mobile applications are becoming increasingly feasible, enterprises must be careful to avoid wasting money on fruitless pilot projects. Here are four important issues to consider when judging whether a mobile application can present a good fit for your enterprise: 

Usability
Compared with most enterprise users, mobile users have to deal with the limitations of PDAs and Web-enabled cell phones--smaller viewing areas and limited input devices. Because they are often face-to-face with a customer or performing some other real-time activity on a job site, these users need a quick response when entering or accessing data. An overly complicated user interface or one that requires too many steps to access or enter data can make the application useless. Before you roll out a mobile application, make sure that its intended users test it in actual job settings.

Integration 
Achieving usability on the front end usually means having strong integration capabilities on the back end. For example, a company may use applications from different vendors for billing, customer data, and logistics. The mobile client may need one or two key pieces of data from each of these applications, all tied together on the same screen. To build such an application, Peter Firstbrook, senior research analyst with Meta Group, says, in some cases, companies will need to look beyond the big enterprise providers. "A native extension may not work unless you are an all-in-one shop," he says. "If a company wants to mine all these systems, then these clients won't help you." Firstbrook suggests that companies investigate middleware from such vendors as Air2Web, Brience, and Infowave.

Online vs. offline use 
If you've ever used a cell phone, you know that wireless networks aren't very reliable. That's a big problem for mission-critical applications. For example, a field-service representative who orders a crucial batch of parts to get a customer's factory back online must be certain that the order arrives at the factory intact. Until wireless networks offer no compromise in reliability, mobile applications will need to be designed for offline use as well. That means mobile users should be able to download and work with all pertinent information on their mobile devices when they can't connect. Once they can connect, they can then upload orders or other data to the home system.

Back-end coordination. 
Creating an infrastructure to provide mobile users with access to enterprise systems makes little sense if the business processes within the enterprise can't take advantage of it. For example, collecting real-time data in the field is pointless if the data is just going to sit in the corporate database, without making use of mobile users' ability to reduce cycle time by reporting and responding to more timely data from the field. When evaluating mobile applications, don't stop at the point where data is received in and delivered from the enterprise database. Diagram the entire process from the field, into and throughout the enterprise, and back out to the field before judging the project's likely effectiveness. 
Probably most important of all, don't be overly ambitious when considering a mobile project. You should move forward only when there is an exceptionally clear and compelling benefit. If "it's high risk, there needs to be a high reward," says Meta's Firstbrook. But even so, the mobile enterprise looks like it's on its way to becoming something more than just marketing fanfare. 

MobileInfo Advisory & Comments:  MobileInfo.Com considers these to be only a partial list of considerations. Surf our website for "issues" in each of the components that comprise a mobile application. These four considerations are a good start for the CIOs..