当前位置:网站首页>Why do you always fail in automated tests?
Why do you always fail in automated tests?
2022-07-05 12:07:00 【Software testing Lao Mo】
Catalog
5、 The existing process is not updated in time
6、 The correct tool was not selected
1、 Preface
Test automation is becoming increasingly popular , It is also a good way for companies to deploy applications quickly and effectively . however , Whether it's making an initial investment or expanding Automation , Test automation will bring some seemingly arduous challenges .
In order to solve these challenges as soon as possible , The following summarizes the five main reasons why automated testing fails . Always pay attention to these problems , As a reference , A guide to test automation for automation projects in the right direction .
2、 Too much Test Automation
One reason why automated testing projects may fail is , Some test tasks should not be automated . Although test automation helps keep up with the release cycle , But automation is not a universal solution to the problem of software testing . Realization 100% Test automation is a highly unrealistic expectation , Companies that try to do this will eventually face greater costs and a series of problems .
Make sure that the person in charge knows what the results of implementing test automation will be , For example, improve product quality by fixing errors or shorten the release cycle by reducing test time . If there is no clear purpose to automate the test , Set them to fail from the beginning . This will also help you identify whether they are suitable for Automation , Or whether manual testing is more suitable for these situations .
It is also important to ensure that automated testing is strategically targeted at the most relevant use cases . Start this process with simple but repeated automated tests . This will lay a solid foundation for building more complex test schemes in the future .
3、 Forget manual test
Another reason test automation may fail is , The test team may not strike a balance between automated and manual testing . As mentioned above , Test automation cannot get rid of manual testing . contrary , The quality assurance team should provide more time and energy to focus on making automation a more strategic decision , Manual testing should be kept in mind when conducting automated testing . The purpose of manual testing is completely different from that of automatic testing , This will affect the design of manual tests , Develop strategies and execution . Besides , Business testers have a unique set of skills , This makes them invaluable to any testing team . They have business knowledge , You can test the application more deeply , And have the intuition necessary to create tests that machines cannot consider independently .
4、 Business logic changes
Automated testing is particularly useful for repeatability testing , But it will take a lot of time and resources in the early stage . Whether it's as simple as testing the login process , It is still as complex as checking the main function , These are excellent test examples that can benefit from Automation .
however , What happens when you want to change the user interface of the website ? for example , What happens when you adjust the login button to make the station more user-friendly ? Fortunately, , There are simple solutions to these problems . Some automated testing tools pass AI Integrated into its platform , Make it easier to test these processes , for example AI How to affect the testing industry , By using dynamic element locators , These tools use AI To overcome such changes in their applications . Back to the login button example , Using a AI The technology tool will be able to find the login button , Despite the size , The color or shape has changed .
Now? , Even the tools used have high accuracy of machine learning algorithm , When these tests still fail , What's going to happen ? An important disclaimer is , Even though AI You can make test maintenance easier after changing the application , But when the tested process changes , It won't help . Back to the login example ,AI Can correctly overcome the changed login button , Without any human intervention . however , It cannot complete the test of adding a new step to the login process , For example, two factor authentication or improving other personal data . To prevent interruption of this test , The tester needs to enter and modify the test to include this additional step in the login process .
5、 The existing process is not updated in time
Starting a test automation test project usually makes participants very excited . After correct implementation , You can quickly grasp how much time can be saved , And how it helps QA The team is more integrated into the software development process , however , Companies that have long adopted automation technology may sometimes be confused by the large number of automated testing schemes that need to be carried out on a regular basis . Quality inspection testers may have some automated tests , These tests are often used when testing new functions , But over time , Other tests may be outdated and cumbersome .
To ensure that the test automation work continues to benefit the team , It is important to spend time optimizing the existing test automation suite . Especially when the test automation operation is more mature , It is important to go back and make sure that the old test cases are still useful , Instead of just focusing on new areas of Automation . This will keep the test operations streamlined , And make it easier for teams to extend Test Automation .
6、 The correct tool was not selected
Another reason automated testing may fail is , Did not choose the right tools to meet the needs of automated testing . When selecting automated test tools , It is important to understand the scope of the tool and whether the functionality it provides is compatible with the team's priorities . Some companies may be looking for solutions that are compatible with mobile applications , Other companies may give priority to tools that can run within their framework .
With clear automation strategies and goals , Choosing the right automated testing tool should be a much simpler process .
Thank everyone who reads my article carefully !!!
If you can use the following information, you can take it away directly :
1、 Self study development or test the necessary complete project source code and environment
2、 Test all templates in the work ( test plan 、 The test case 、 Test report, etc )
3、 Classic interview questions for software testing
4、Python/Java Automation test practice .pdf
5、Jmeter/postman Interface test full set of video acquisition
I personally sorted out some technical materials I have sorted out in my software testing career in recent years , contain : e-book , Resume module , Various work templates , Interview treasure , Self study projects, etc . If you encounter problems in your study or work, you can directly join the group and ask , There will also be great gods in the group to help solve , If you need anything, please give it to me personally. Thank you .
边栏推荐
- C # implements WinForm DataGridView control to support overlay data binding
- The most comprehensive new database in the whole network, multidimensional table platform inventory note, flowus, airtable, seatable, Vig table Vika, flying Book Multidimensional table, heipayun, Zhix
- [cloud native | kubernetes] actual battle of ingress case (13)
- Mmclassification training custom data
- pytorch-多层感知机MLP
- [calculation of loss in yolov3]
- Course design of compilation principle --- formula calculator (a simple calculator with interface developed based on QT)
- 一次生产环境redis内存占用居高不下问题排查
- 什么是数字化存在?数字化转型要先从数字化存在开始
- Is investment and finance suitable for girls? What financial products can girls buy?
猜你喜欢
iTOP-3568开发板NPU使用安装RKNN Toolkit Lite2
Pytorch weight decay and dropout
Simply solve the problem that the node in the redis cluster cannot read data (error) moved
pytorch-权重衰退(weight decay)和丢弃法(dropout)
Uniapp + unicloud + Unipay realize wechat applet payment function
13.(地图数据篇)百度坐标(BD09)、国测局坐标(火星坐标,GCJ02)、和WGS84坐标系之间的转换
Four operations and derivative operations of MATLAB polynomials
[calculation of loss in yolov3]
多表操作-自关联查询
【pytorch 修改预训练模型:实测加载预训练模型与模型随机初始化差别不大】
随机推荐
Hiengine: comparable to the local cloud native memory database engine
Reading notes of growth hacker
【上采样方式-OpenCV插值】
What is digital existence? Digital transformation starts with digital existence
What is the difference between canvas and SVG?
Matlab struct function (structure array)
Want to ask, how to choose a securities firm? Is it safe to open an account online?
Course design of compilation principle --- formula calculator (a simple calculator with interface developed based on QT)
byte2String、string2Byte
Pytorch weight decay and dropout
C # implements WinForm DataGridView control to support overlay data binding
Redis集群(主从)脑裂及解决方案
想问问,如何选择券商?在线开户是很安全么?
abap查表程序
Redis cluster (master-slave) brain crack and solution
The most comprehensive new database in the whole network, multidimensional table platform inventory note, flowus, airtable, seatable, Vig table Vika, flying Book Multidimensional table, heipayun, Zhix
July Huaqing learning-1
liunx禁ping 详解traceroute的不同用法
Programmers are involved and maintain industry competitiveness
Riddle 1