Why automated testing is hard to scale

It has been ten years since I first came into contact with automated testing in 2005. Looking at the companies around me, very few companies actually implement automated testing. For most companies, testing is still at the point-and-click stage. Testers are usually the acceptance deliverables without being involved in the entire software development cycle.

Why automated testing is hard to implement

Why automated testing is difficult to implement, I think there are several issues that hinder the popularity of automated testing. In fact, there are still many people who understand automated testing tools. It is not a lack of technical talents that automated testing is difficult to implement. Load Runner, QTP, etc. are used by many testers, why do they abandon these tools and use manual testing?

90% of tests are still functional tests
Many testers have no development background
Testing role, not throughout the software development cycle
Various issues hinder automation scripts
The labor cost of testers in China is too low

With the development of technology, the diversity of software is not limited to GUI based on CS structure, but based on BS browser WEB UI. For example, the current Android system, Apple IOS system, Microsoft's Windows Mobile system and so on. There are also some non-human-computer interaction interfaces, various protocols/interfaces, such as json, bson, xml-rpc, soap, mq (message queue) I think these should be included in the scope of automated testing. This requires the tester to have certain development capabilities, and the test of the above content requires extensive technical knowledge support.

I think a senior test engineer needs to have the following abilities

sniffer use
gdb use
Learn about various protocol families
infiltrate
HTML/CSS/Javascript
database etc.

As far as WEB testing is concerned, the content involved is too extensive, from browser->WEB server->APP server->cache->database, through various proxies, load balancing, distributed file system and so on.

It is very difficult to configure such a test environment. Fortunately, we can use automated operation and maintenance to do this.

What's holding back automated testing

Various UI effects
verification code
Browser support
Third-party plugins (Flash, ActiveX...)
technical closure

互联网的快速发展 Load Runner, QTP 等等软件，我认为已经跟不互联网的快速了，他们仍然按照传统周期发布软件更新。而互联网需要的是快速变化，互联网应用程序开发者，需要体验更多的创新功能，软件软件发布周期至少一年一个版本。真的太慢了。

互联网不断加入的新技术成为了自动化测试障碍，传统软件无法支持这些新技术，甚至向微软这样的企业技术跟进都显得不给力。

Windows Automation 3.0 是非常高大上玩意，但是你在Microsoft官网能找到的资料，少之甚少，我不知道微软的目的何在。

只有 Load Runner, QTP 这些功能与微软又合作，才能拿到Windows Automation API。

中国测试人员的人力成本

测试人员的薪水在开发团队中应该是处于中下等的。与高级程序员，软件架构师是有很大差距的。这也造成了自动化测试难以实施的原因。

我们需要从高级程序员，软件架构师转测试的高级测试人员。

我们需要黑客级的测试人员！！！