This is a difficult question to answer without more information about the three products. I would suggest you sit down with each team to understand their testing needs and move from there.
Integration of QA in work processes
In the company where I work there are 3 products: software, application and site - each with a different QA process. The question is how would you suggest integrating the QA process in different scenarios? Because there are tasks in which the QA man checks the task and there are those who "created" the task and checked whether it was indeed done at his request.
I look forward to your comments on this subject.
I am assuming that the three 'products' you are talking about are integrated into one deliverable. That is to say, that you are delivering a product with some desired utility to an end user group. I understand the 'site' component, but I am not sure what the difference is between 'software' and 'application'. If my assumptions are correct, then there should be some type of master testplan (or some type of guideline) outlining all the processes/strategies included in the the QA certification of the product. All groups should be adhering to this plan/strategy. Each group; Business, Development, QA etc. should have defined tasks with regards to your company's implementation of the SDLC. For example,the business might be required to identify the product requirements and perhaps assist in the design of any user interfaces. Development may be required to perform unit tests or code walk throughs. The test group may be required to develop functional and integrated test suites and associated automation.
The point here is that for the QA process to work properly, all stakeholders need to perform their specified tasks. While there is often some overlap, each group should have their own unique tasks assigned. If you are unfamiliar with the industry standards for SDLC processes, IEEE has some good documentation to get you started. You can then combine that with any of the Agile process your company currently uses.
If however, you are saying that there are three distinct products being delivered, then that is another thing. From this perspective, aligning different projects in terms of QA would call for a company level decision requiring all projects to follow some defined set of standards. Something like the (SEI) Capability Maturity Model.
I hope this helps a bit.
I think it would be easier to answer your question If I get to find that the software, application, and site belongs to one entity and have a shared purpose. Like if it is anything similar to an eCommerce business where a website and application can all be used to lead the business operations.
If that is the case, I believe they are all likely to share features and functionalities which means the fundamentals of testing such as unit testing, functionality testing, usability testing, UI testing can all be integrated. On top of that, I would still recommend automated regression testing to ensure that the application and website both offer similar performance.
However, applications would need you to test for different operating systems and websites needs to be tested for offering cross-browser compatibility.