You've completed your VDI proof of concept -- that is, you've chosen a VDI product to use based on your own audition...
process of all the different options available -- and now you're ready to roll into the pilot phase.
Though you may only be dipping your toes into the VDI pool to get a feel for how it will perform, in some ways it may still feel like you're jumping in head first. On the technology side, your proof of concept has probably made you confident that you've picked the right platform. But for this project to succeed, your users have to buy in, too. Choosing the right VDI pilot users is crucial.
So how do you choose the most appropriate users to get virtual desktop infrastructure (VDI)? Some IT administrators take the dartboard approach, picking random people from all around the organization. Others draw names from a pool of employees who are known to be savvy pilot users because that group can better tolerate small issues before pushing the panic button. Still other admins choose a single department and roll VDI out a department at a time.
Each of those is a very common way to pick VDI pilot users. But to get the best idea of whether VDI is right for you and going to work, you have to take the randomness -- or predictability -- out of the pilot.
Those words may seem to contradict each other, but think about it: If you randomly deploy VDI, you might not hit your use cases and get feedback from people who don't have the problems you're trying to address. And if you deploy VDI to a group of tolerant users, you may not get the proper feedback on things that would affect other, less intuitive users.
The real way to choose the VDI pilot users is to look at the goals that caused you to look to VDI in the first place. For instance, if your use case is mobility, then make sure your pilot users are mobile workers. From there you can spread it around to all-day mobile users, traveling users, and occasional mobile users. If your use case is security, focus on users who have or are susceptible to the security issues you're trying to resolve. If you're trying to deliver Windows desktops to students, don't deploy the pilot to teachers.
Of course, if you're trying to accomplish something more fundamental, like reducing down time, enabling disaster recovery, using single images or stretching your endpoint, OS or application lifecycles, the random sampling approach might be useful. Just don't go to it by default. The more thought you put into the VDI pilot users, the better.
There are a million reasons to do VDI, so choose the users that fall into the bucket of problems you're trying to solve. At the end of the day, the most important thing is to make sure you isolate your goals. If you get to the pilot phase and can't determine which users fit your use case because you can't figure out your use case, step back and look at why you're doing VDI in the first place. Perhaps another aspect of desktop virtualization is more appropriate for you, or perhaps you got caught up in the buzz. Finding that out now is a lot better than learning it halfway through a production rollout.