When preparing a presentation for TechEd Amsterdam on mobile lately I was trying to wrap up my session on operating mobile landscapes on two slides. I wanted people to remember these things when walking out of our workshop. I came up with the following list that I consider a good summary of the things to keep in mind when planning a mobile project and when operating the resulting mobile landscape.
Here’s my top ten things to keep in mind …
1. Every day tasks are more error prone in Mobile Environments than in traditional IT environments
Sure, the complexity is much higher due to the fact that there’s usually more devices, that you have client, middleware and backend, that there are different technologies like Java on the client and ABAP in the backend involved …
2. Mistakes are hard to fix – since they might require rolling out the software to possibly thousands of clients again
If you have to fix something it means more work than in traditional environments because of the distributed structure of mobile environments.
3. Therefore a good planning is required and strategies on how to deal with potential issues have to be developed
You better plan ahead and have something ready before a problem occurs.
4. Invest a high percentage of your time and resources (up to 30 percent) into rollout and support
A good support and rollout strategy enables you to react to issues in a timely manner.
5. Extensive end user education is recommended since Mobile is a new technology for end users
Education of end users helps you to prevent issues before they occur.
6. Ensure the support is able to access devices in case of problems either directly or by remote login
This speeds things up a lot. It’s only possible for Laptops and Tablet PCs though.
7. Device Selection is crucial. If you choose the wrong hardware or save on the hardware you might pay the price later
The selection of devices is actually the MOST important point. You should have a look at a number of devices and choose one that fits your needs. Spend some time on this especially when you think about getting PDAs.
8. Involve your end users as early as possible. Especially when selecting devices
The success of your project primarily depends on how your end users accept your package – meaning the devices and the software. Usually possible issues come up early in a project when you involve end users and, even more important, end users appreciate being asked. After all your end users are the people that you want to help doing their jobs …
9. Think twice when administering devices. You might delete a device that is still in use. Or, worse, you could delete a large number of devices that are still in use
Small things can have large effects. So THINK when doing things like deleting data from the middleware …
10. Make sure that your sizing fits your requirements
You would like a solution that is comfortable to use and offers enough performance. So make sure that both your device and your middleware are able to handle the expected amount of data.
And an eleventh point to keep in mind that is even more important than the other ten: Mobile Projects Pay Off. You might have to plan things more carefully than for a traditional IT project but you will be rewarded by a much higher return on your investment since most of the infrastructure and software (Backend) is in most cases already payed for before you start the project.
And when do you get the chance in a traditional IT project to deliver software access to 10000 users at a time that had no access to your company’s software before ?
As I said: mobile projects do pay off 😉