Story: A Former Mentee Failed Test Automation with Protractor. Part 2: Reflections and FAQ
The price to pay for not following a proven winning formula.

Continue from Part 1.
Reflections
Real test automation and Continuous Testing are more profound when you are aiming big (and with existing legacy test scripts). Surely, F would face some challenges the next day but realized he couldn’t ask for more free mentoring. Therefore, giving up was inevitable.
There were several other stories like F, some of my mentees failed on
- using Gherkin syntax, i.e. Cucumber / SpecFlow.
- using Java/C# as the scripting language.
- running automated E2E tests in CI servers such as Jenkins/Bamboo/TeamCity/GitLab.
- using auto-retry in ‘framework’ instead of in the CT server.
- …
Among them, F’s story is the most convincing one. He received the proper real test automation and CT training/mentoring at the beginning of his IT career. Quoting his words: “I will take well with me all my life, and they are skills that put me years ahead of my peers”. It is true that my mentoring helped accelerate his developer career. However, he forgot what I often said during mentoring:
“Do hands-on automated UI testing every working day, no matter that you work as a software engineer or tester.” — Zhimin Zhan
Of course, strictly every working day is not possible, but you get the idea. I received similar advice from my mentor, a world-renowned programmer, back in 2005. I followed that advice, and I will be grateful for that forever.
FAQ
1. What is the best step for F to take in this case?
First of all, moving away from Protractor (to raw Selenium + RSpec) as it is being deprecated.
Put aside the company factors over which he might have limited control, if there is an urgency to get the automated test suite back to a working state, he should engage me as a test automaton coach to re-implement them quickly in raw Selenium WebDriver with RSpec. It won’t take me long, maybe one to two weeks to cover the core business features if the testing infrastructure (such as reliable servers, good test data defined, ..) is good. Then he can take over (which I believe he is capable of) with occasional mentoring support from me.
2. What if there is no possibility of engaging ongoing external coaching?
Successful test automation is still achievable by following my books:
- eBook: “Practical Web Test Automation with Selenium WebDriver”
- eBook: “Practical Continuous Testing: make Agile/DevOps real”
and my Medium articles.
Do it slowly, but make sure to run all tests often, and all tests pass (or are valid) every day. Slow and Steady wins the game. Refer to my 35-Word Functional Test Automation Strategy.
If there is team involvement, a fast track way is to attend my one-day training first.
3. How can you be so sure that your formula is 100% correct?
At least from F’s point of view, my formula was correct, he did see (and got involved in) it years before, and the fact he reached out to me for help after so many years confirmed that.
Also, there is no dependency and no cost with raw Selenium + RSpec (my tech stack). If there is a ‘better’ option, they are free to convert. F and/or his colleagues went for a ‘better Protractor’, which turned out to be bad (and worse, their chosen framework was deprecated).
That is why an objective test automation strategy is important, like this one. I have been following that since 2007. It is totally fine to try alternatives, but don’t lose the goal.
There are numerous examples going for other bad frameworks such as TestCafe, Cypress and Playwright. Read this well-received article: An IT Graduate’s frustration with a Fake ‘Senior Test Automation Engineer’.
The problem here was lacking objective assessment and kept going on the wrong path. From a different perspective, this company had never done successful test automation before, why not follow my proven formula first? If they did, the results could be seen in just days. The blind decision to go for JavaScript was costly. Once it happened, companies are often too embarrassed to acknowledge and change.
---
This article was originally published on my Medium blog on 2023-05-27.
Related reading:
- My eBook: “Practical Web Test Automation with Selenium WebDriver”
- My eBook: “Practical Continuous Testing: make Agile/DevOps real”
- Test Automation Camel, a metaphor that explains why most test automation attempts failed?
- Benefits of Test Automation and Continuous Testing series: Executives, Managers, Business Analysts, Developers, Testers and Customers.
- Testing Pyramid Clarified
- Which Selenium WebDriver Language Binding is the Best?
- Crazy Web Test Automation: “Freedom Is Slavery”
About the Creator
Zhimin Zhan
Test automation & CT coach, author, speaker and award-winning software developer.
A top writer on Test Automation, with 150+ articles featured in leading software testing newsletters.



Comments
There are no comments for this story
Be the first to respond and start the conversation.