FAQ – Whats the ideal percentage of test coverage?
 |
There’s a very interesting thing about test coverage in the QA industry. It’s not exactly well-defined. Because when you’re talking to engineers, they mean one thing for test coverage; however, QA can mean a completely different thing. So for engineers, test coverage usually applies to unit tests indicate aligned coverage of lines of code. So how many lines of code are covered? You can measure that precisely, and you can calculate with precision. However, then it goes to QA because QA often deals with applications on a business level, and it deals with end-to-end tests rather than unit tests for QA Test coverage means how many business scenarios that had been outlined are covered.
As you can imagine, it’s not a precise enough function because it’s the function of how much time was invested into outlining all-important business scenarios to cover because you usually it’s almost impossible to digest all of those. It usually covers only the most important ones. So in that regard, they believe that the right amount of test coverage is 100% of what can become, of course, various scenarios that account thought separately currently recovered, but what out of what can be carried this should be the reason why people think it shouldn’t be 100% is because we used to do with older technologies, like Selenium, with Selenium’s really a huge pain and effort to build tests, which is technically a struggle to get to high coverage percentage and is almost impossible to achieve; however, with testRigor you can achieve 100% test coverage within a month by copy-pasting your own test case written in English into testRigor and making executable using plain English as testRigor plain English execution system. This way it will be pretty quick and easy to automate everything you have, whatever can be automated. Moreover, they believe that all of those test cases must be automated because it improves the speed of business and if there are no issues with test maintenance anymore, as it would be with Selenium, you should cover all of your end-to-end tests scenarios because we believe end-to-end tests are the most critical tests, you can run them because they’re validating actual business functions that your application functions from the end user’s perspective end-to-end and your users can achieve what they set out to accomplish in your application.
