Thread: Advantages/Disadvantages of manual software testing?

Hi Nagesh

I have posted ur question to the forum pls check.

Can you plz tell me that when we go for Manual testing and
when do we go for Automation Testing.
From where does we start Manul Testing and end.
and
From where does we start Automation Testing and end?



Manual testing is the testing done without the use of any automation tool

Automation testing is the testing done with the help of a tool.

Manual testing start once the testing team receives a build from the development team. (the zero level testing starts here).

Automation testing this depends on the requirements indicated to perform testing. Not all test cases need to be tested using the automation tools. some test cases have to be tested manually.

Automation testing takes place when we have to perform functionality testing using multiple data or we have to perform regression testing.

Start and end of the automation testing depends on the factors like length, duration, cost of the project, risks of the project etc.,

It’s simple,
Manual Testing means, we have to input and trigger actions in the s/w application ourselves , while in automation testing, the tool takes the ownership of these actions .

If it becomes tedious and time-consuming to manually input data and trigger actions in the s/w build, we go for Automation testing.

Manual testing or Automation testing starts with the process of writing test cases and ends with the successful completion of all the test cases.

Explain disadvantages of manual software testing and how we can avoid those pitfalls without switching to automated testing?

Manual testing could be,

1) Time consuming - we may follow BVA and ECP techniques to overcome that,

2) Tedious - we may distribute tedious test cases equally among team members or avoid executing repeated test steps of 2 different test cases.

Originally Posted by sutnarcha

Manual testing could be,

1) Time consuming - we may follow BVA and ECP techniques to overcome that,

We use BVA and ECP even we are doing automated testing also.So, it's desirable and required to do so.

2) Tedious - we may distribute tedious test cases equally among team members or avoid executing repeated test steps of 2 different test cases.

If there is a test team, work is always divided into members as per exp. and qualifications.

Disadvantages:

Can be time consuming
Supporting material not always available
Requires significant human thought and skill to be effective!
__________________


Boundary value analysis and equivalence class partitioning. Equivalence partitioning - it is a black box testing technique and it can be defined as "dividing the input domain of data into the class of intervals and from which we can prepare test cases. Ecp: if an input specified member of sets, then one valid and one invalid value included boundary value analysis - it is also a black box technique major stress on boundary conditions, as the most of the bugs reside at boundaries. Bva: if an input is bounded on 20 and 30, the test case should included the values 20, 30 and just above and below of 20, 30 respectively .

These questions are answered quite a few times in Geek. Search for them.

we go for automation testing when the nature of the testing is repetitive. Testers get bored when they are exposed to, testing a part of the application again and again finally leading to frustration. carrying out manual testing in this case would result in high consumption of both cost and time.

Another situation where one should go for automation testing is in performance testing. while carrying out testing for web applications, such as how much load a system can bear at a time and system's response time, system should be able to emulate thousands of users accessing a server simultaneously, which is literally impossible when comes to manual testing.

Basically in other situations automation testing helps in speedup the process of testing in which case atleast one round of manual testing should have been done.

Dear Lata,

For Automated Testing, scripts are written in whatever the tool used may be. This scripts needs to be tested.
The script does what the script writer asked the system to test. It may not necessarily do what the script writer wanted the system to test!
This could result in errorneous reporting of defects (which are not actually defects in the AUT but a defect in the script itself) or could even not report defects in the AUT since somebody missed to include that in the script.

Therefore, there is a fair amount of chances that, the scripts will fail as the script itself is a software and software does fails.

In manual testing since the human brain is directly involved, the output of the testing can be analyzed and we may come to some logical conclusion.

While testing we need to analyse the AUT and come to the best possible approach of testing and i believe it should be a healthy combination of Manual and automated testing with all other static testing strategy.