Implicit indicators for evaluating the advantages and disadvantages of automated testing

　　 About evaluating the advantages and disadvantages of automated testing , In addition to common indicators , There are also some indicators that are not easy to talk about with data , This is called recessive index .

　　 Hidden indicators mainly include ： Automated maintenance costs 、 The running cost of Automation

　　1 、 Automated maintenance costs

　　 For the same business , Different automatic test implementation schemes , The corresponding maintenance costs may be vastly different . indeed , Automated maintenance costs , Affected by business maturity 、 Iteration speed 、 The degree of project specification affects , But consider the following situations , How about your maintenance cost ：

　　 Added some logic ( Such as , Interface / service / application ), Automatic maintenance of new parts , How long do you need ;

　　 Deleted some logic ( Such as , Interface / service / application ), Automatic maintenance of deleted parts , How long do you need ;

　　 Modified some logic ( Such as , Interface / service / application ), Automatic maintenance of modified parts , How long do you need ;

　　 When the iteration speed of the project accelerates , Accompanied by logic of addition, deletion and modification , Can your automated script keep up ? Actually , This is a severe test that many teams will face . A business in rapid development , Every business test 、 When returning , May want to give up automated testing , Instead, perform the test manually . because , Automated testing also requires constant debugging of automated code , The probability is too late for this test , It is not as efficient as direct manual testing .

　　

　　​

　　2、 Automation operation cost

　　 The cost of automation here is , From want to implement automation to the end of implementation, we need to meet the ability & Time cost . The general automatic operation process is as follows ：

　　1) Create some automatic execution conditions . such as , Find running data , Set up the running environment, etc , If this step is not automated , It takes manpower & Time ;

　　 for example , Automation can only " Walking on one leg ", That is, it only realizes semi automation , It didn't happen 100% Automation , Before running / in / After that, people may need to participate .

　　2) Execution automation . Here is mainly the time required for automatic operation , The longer the time , The longer the waiting time . Maybe you'll say , When automating execution , You can do other things , There is no need to wait for the execution to end . But now that automation has been implemented , It must be like manual testing ," immediately " See the execution result , Get timely feedback , To avoid switching back and forth between different jobs .

　　3) Verify automation results . The verification of general results is included in the previous step , However, it is not ruled out that some verification still needs to be done manually check The situation of . This situation , In fact, it also belongs to a form of semi automation .

　　4) Automated failure troubleshooting . All kinds of reasons , Will lead to the failure of Automation , such as , Data problems 、 Environmental problems 、 Automatic maintenance is not timely 、 Third party problems, etc . After automation fails , Can you give a clear reason for the failure , Even according to the result of automation failure , Directly locate the failure reason , This is in place , It will make you love Automation .

　　 For more information on testing techniques, please pay attention to ： Shenzhen duosurveyor software and Technical Service Co., Ltd