| The order in which Martin would roll out the hardware and software was quite straightforward. As each trading unit was distinct, it was easy to schedule and deliver the installation of both the hardware and software. Hardware could be installed at any time in the future and activated after the software was installed. Jason had a challenge in front of him. Since the biometric project was tied to the digital certificate project, the order of rollout was ad hoc . Any employee could apply for and receive a digital certificate on-the-fly . That meant that the hardware needed to be put out as early as possible. |
The Methodology Once the automated install is ready and the pilot user selection is complete, it is time to get the pilot deployed. The deployment order of the pilot needs to be done according to the enterprise's needs. This ordering of the pilot rollout may be done in the following ways: -
By business line ” The pilot community should cover as many types of users as possible to give an overall representative sample. The order of deployment does not have to cover all parts of the user community at once. It may make sense to deploy along business unit lines. that is, a certain business unit may be easier to do first. It may also be required that the entire pilot population of that business unit be completed at the same time. This could be for reasons of compatibility with each others' business processes, to support a free or hot seating model, or because it is simply a good time for the unit to be deployed. -
By level of effort required for installation ” It may be known from the POC that one particular desktop build or business unit will be easier to get working than another. In this way, the pilot would be rolled out from the easiest to the hardest. This allows the pilot to build momentum, confidence, and success. Then, when the pilot deployment hits difficulty with more challenging deployments, there will be confidence in the solution, and troubleshooting and resolution will be easier. -
By risk of failure ” Like level of effort, the risk of failure of the biometric system or the desktops of the pilot users must be evaluated. Users for whom the risk of failure is high within the context of the business or who would have a high downtime cost should be deployed last. By that point, most issues with the pilot deployment should have been identified and addressed, and additional support resources will be available to help support these high-risk users. Also, it goes without saying that the executive users should not be done too early. Senior executives quite often want the latest software and hardware before anyone else. In doing this too soon, the additional resources required to support the executives would impact the ability to deliver support to the rest of the pilot community. Like other riskier users, they should be done near the end of the pilot. The final decision on how the pilot will be deployed may be influenced by all, some, or none of the above. Also, it is important to be flexible and change how the pilot is deployed as it moves forward. It is of no use making mistakes if you cannot learn from them and adjust. Thus, the deployment plan is only a plan; as such, it may need to be changed and updated as it moves forward. |
