Keyword-driven testing: reasons to use or skip. Part 2
In the previous post we had a look over keyword testing that has a somewhat dubious reputation with QA consultants and covered some steps of the comparison we made. This I`d like to provide you with the test results we`ve got.
Several charts below illustrate our findings regarding last week of experiment. One should also note tests were executed daily.
The first comparison had respect to test execution time. Keyword driven assumes that some additional logic of parsing keywords works at the background. The impact on test execution duration was mainly under our focus.
The results indicate test execution time is comparable and we may apply Keyword driven in future without additional efforts/hardware for optimizing test duration.
Efforts required for analyzing test results became another comparison point here. In fact it shows the similar picture. Despite keyword driven required a little bit more efforts to support tests, we got the same trend there. The reason may lie in the sphere of business logic. This part of measurements should be developed further.
Another point is test support time. We may see that both solutions are comparable by this indicator as well.
As for the test development efforts – manual QA Engineer was able to develop test case description using existent keyword. Taking this into consideration we may consider spending to be effective comparing to current solution.
Let’s sum up the outputs
The first conclusion we are able to mention is that Keyword driven has no contraindications for use and POC stage showed positive results. On the other hand based on our measurements we see no extraordinary results or substantial breakthrough.
We started the measurements after test development efforts of the new cases became comparable and we’ve got pretty the same characteristics for critical automation process on both solutions.
So, based on this research one can suggest using keyword driven when you are going to cover hundreds of test cases in long term projects with a business logic which assumes high level of keyword re-usage.
Having this as a consequence we decided to apply Keyword driven practice to several upcoming projects.