A lot of developers and a minimum of testers - is this a familiar picture for you? Sooner or later, two problems arise with this approach:
- New functionality or bug fixes begin accumulating in the QA task pool. Developers are periodically stagnant.
- Testers have very little time left to develop autotests. Most of the work with the code is occupied by fixing existing tests, which broke during the latest functionality changes already covered by tests.