I am working as the QA Manager and we are using QC 10.0 for our defect Tracking and Test Planning and execution. We have multiple releases in a year and we create one project for each Release. So if we have 4 releases for a year, we end up creating 4 Projects for that year. The QC Administration part is maintained by a outsourced vendor and for setting up the Project for one particular release, we have to pay 3000$.
The Management does not want to pay the $3K for each and every release. Rather they would like to have one Project at a high level and have sub folders for different releases.
I would like to understand how feasible this will be and what would be the challenges involved. What would be the factors that we need to consider before going ahead with such implementation.
Some factors to consider:
1. How do you want to report on the data? This will impact how you organize the data in a single QC project. If you want to be able to report on data per Release, you need to factor that in when organizing the data. The Release Management module can help you "tag" data to support reporting by release.
2. Who should be able to add/change/delete data? Is this different per Release? If so, maintaining the requisite permissions to keep folks from accessing data in other Releases could be complicated.
3. Do you use the same customizations in all projects/Releases? Think about user defined fields, customized workflows, customized lists. If they are different per Release (per project), trying to accommodate those in one QC project could be difficult.
If your QC projects are identical from one release to the next you should really renegotiate the fee. It takes about 5 minutes to create a new project from an existing one.
If you are not making frequent changes to your customisations and you don't have huge data sets, managing multiple releases within a single project is fairly straight forward. Keep in mind that HP have tried to make it easy to manage multiple releases from a single project.
One of the clients I worked for had manual test cases in the 10s of thousands and customisations were rarely consistant from release to release so the manual team had a new project for each release. My team on the other hand maintained the automated test cases and I had complete control over customisations so we only ever used one project.