For enterprises, deploying wireless applications to employees often means confronting a long and daunting list of concerns. But experts say that keeping an eye on the business case and remaining practical can help ensure successful deployments.
Stephen Poppe, chief information officer with Roto-Rooter Services Inc., a subsidiary of Cincinnati-based Chemed...
Corp., knows all about the problems that can crop up in a wireless deployment. He recently launched a wireless service for Roto-Rooter's 1,500 field technicians, who are spread across 70 locations in the U.S. and Canada.
The key, Poppe said, is keeping the project well defined during the research stage.
That means determining whether a wireless application is needed, or whether a mobile one would do the job just as well, said Phillip Redman, research director with the Stamford, Conn., research firm Gartner Inc.
Redman said that, often, project managers think they need a wireless application, when in fact they only require devices that workers can use in the field to input data that can be synced up at the office later. Choosing this latter route, as opposed to one that requires supporting full-fledged wireless applications, will make a deployment easier and less expensive, he said.
Making the leap
For those enterprises that truly need wireless connectivity, the key is to keep the applications simple and focused on the company's business requirements, Redman said.
Some wireless devices come with lots of bells and whistles, such as color screens and a range of other capabilities. However, the ability to deliver pertinent, real-time information to users is much more important. Wireless connectivity is about giving people the information that they need when they need it, Redman said.
One of the first wireless deployment issues that crops up is how to provide wireless service, particularly among the muddle of air interfaces in the U.S. Today there are at least a half dozen in use. General Packet Radio Service (GPRS) and 1xRTT have emerged as the most popular specifications. These are 2.5 generation air interfaces, meaning they are packet-switched, always-on networks that are geared toward data transfer. 1xRTT, part of the CDMA family, is faster right now, but GPRS is the standard for most of the world.
But the air interface is much less important than the availability of the service, said Peter Firstbrook, senior research analyst with Stamford, Conn.-based research firm Meta Group Inc. The newer technologies, such as GPRS, are less likely to have a large footprint. They may offer higher speeds, but the service area is likely to be small. Often, older, slower technologies such as CDPD have much better coverage, but the drawback is that they are slowly being phased out, he said.
In addition, Firstbrook said, you can't rely on a carrier to be accurate about its quality of coverage.
"The carriers are really bad; they do everything they can to make it sound like they have good coverage," he said.
The best way to determine the quality of coverage is to simply drive around the coverage area and see whether the service works where it is needed, he said.
Poppe used a similar approach to determine the best carrier to go with. His technicians used wireless service from Nextel Communications Inc. for voice. Since the company's data service is laid over its voice service, he knew that data coverage would be as good as voice.
Managing the journey
Device management for mobile applications is also a challenge, particularly for field force applications like Poppe's. Since his technicians only visit the office once a week, updating software through syncing was a bit risky. Any problem updates would not get fixed for a full workweek. By working with Nextel, he was able to set up a system that updates software wirelessly during off-hours, when phones are most likely in a coverage area.
Mobile devices are also much more likely to be lost or stolen than desktop computers, and theft can put important data at risk, Firstbrook said. Vendors such as Hewlett-Packard Co. have developed encryption software for Microsoft Corp.'s Pocket PC device that helps to secure data, he said.
Perhaps the most important consideration, especially when budgets are tight, is accurately determining the return on investment. Wireless e-mail alone can cost as much as $1,000 per year per user, said Firstbrook, and it is not easy to estimate a dollar figure for the gain in productivity.
Some companies may be able to justify a wireless application implementation to management by claiming a boost in productivity but, for many, that is not enough. Poppe knew his company's rigorous accounting department would need more proof. Though his field service workers were able to save time -- and often able to make more stops in a week than before -- that was not considered a "hard" return. He had to look elsewhere to make his case.
Soon, he realized that one of the benefits of the wireless service was moving from a paper record system to an electronic system. With fewer hands on the data, there are fewer errors, and less time spent on data entry.
Another boon was the implementation of a mobile credit card swiping system. With that, the company saved on the increased commissions that the credit card company takes for old-fashioned carbon copy credit card receipts. As a result, he could point to hard dollar gains from the application.
The key with calculating ROI, he said, is to avoid raising expectations too high. There are plenty of futuristic possibilities with wireless but, he said, if an enterprise focuses on those and they do not deliver, a perfectly good application will look like a failure.
FOR MORE INFORMATION:
Read more of our Special advisory: An introduction to enterprise mobile computing
Browse our Topics on mobile management issues
Learn how to plan today for future wireless success