Teeter-totter in testing

In automated software testing special software are used to manipulate the execution of test cases and compare actual outcomes to projected outcomes. However one can automate some repeated tasks that would be difficult to perform manually. Here 100% automation of testing phases is neither theoretically possible nor practically. Therefore user intervention is required to certain stages where automation is difficult or impossible. In arena of automated software engineering, a hot question is measure that what should be percentage of manual portion an automated system for testing. It is also pertinent to research that what phases of software testing can be automatic and what cannot be. This type of discussion leads us to comparison of current research work in the field of automated and manual software testing. This type of comparison will also be an indication of appropriate amalgamation of automated and manual portion of testing. In this paper, the focus of research is to identify relationship between automated and manual testing and to find proportion in both methods to reduce the risks of failures and bugs escapes into automated software testing based products. Furthermore, we aim to investigate that where one can end up with manual testing and start automation and vice versa.