How to deal with criticism with being a tester
A while back, I had the pleasure of doing an online interview with Viv Richards and Graham Ellis around my contribution to the book ”Around the world with 80 testers”. My contribution is based on a talk called ”My journey from dev to tester” and it is about my experience and learnings from that career switch. It was a lovely chat, even if we got some zoom bombers, and a great opportunity to further explore a few of the areas that were really condensed in the book contribution.
Afterwards, I was approached on Linkedin with a question and I would like to share the question, and my answer, here on my blog. (with permission of course!)
The question was:
I’m a junior tester with a software engineering background and I would like to ask for your advice about how to deal with the criticism of being a tester. I’m the only one from my school to go against the flow. Everyone is pursuing artificial intelligence or data science or development and they’re considering testing to be an easy job
Interestingly enough, this is something I had been thinking about a lot after the interview. You see, one of the questions I got was how I experienced moving from a role with ”more status” to one with ”less status”. I’ll get back to that part a bit later.
First, here is the answer I wrote on Linkedin:
Ok so for one: learn to accept it, not take it personally.
Testing is, has been and will be, viewed as ”lesser” than other parts of software development.
We both know it would crash and burn without us but ut _is_ a lot of magic work. What we are doing in invisible and they only miss us when we are gone.
As for the ”easy job”, well… How hard is AI and Data Science? It’s just logic. It can be as easy or complex as you make it. You can use other, prepared, tools and models and ”just” analyze the outcome OR you can build extremely complicated state-of-the-art stuff.
Testing is just the same.
Following a script and confirming 1+1 is still 2 is easy. And that is probably what they see as testing. Especially since a _lot_ of testing out there is done by really non-skilled or non-engaged testers.
Plus the really great testing _looks_ easy. My best testing work has been inserting the right question into a discussion and saving the company hundreds. of hours of writing the wrong software, that is not something that shows. I know I did it. The team probably already forgot about it.
Testing can be done scientifically. It can be done as random pressing buttons. It can use AI and ML. It can be done with very sophisticated data analyses. It can be writing extremely complex automation tools. It can be psychology.
So sure. It’s just as easy as front end development, design, UX, AI, ML, back end development, embedded programming, management or building a rocket.
Connecting that back to the question I got in the interview, about status.
I never understood that when I did the switch. In my little bubble, testing was for sure seen as something boring but never easy. Nor did my colleagues treat me as someone lesser than them, rather the opposite! For me, moving into testing was a power move. It was a way to get more influence, a way to open up new career opportunities.
Programming, for me, had become a routine. I found no(/few) challenges and my earlier tries at showing interest in new roles (architect, DBA etc) had been shut down.
When I found testing, I found the challenge I had been missing.
Yes, to be fair: I came across the notion later on in my career. A lot. And I am very happy I got to experience the first years amongst friends who trusted me. And yes, having 10+ years of programming behind me of course gave me a different status than if I had started out fresh from school.
Summing it all up, my view is:
A lot of the programming going on out there in the world is not very hard. It is doing the same thing, solving the same problem, over and over again. Of course, we also have programming that is so complex it makes my brain melt even trying to understand it. And everything in between them.
Design can be brain numbingly easy. And it can require years and years of studying human behaviour, analysing data, doing complex experiments. Applying science to steer users to behave in the way we want them to. And everything in between them.
ML/”AI” can be downloading a ready-made model, following instructions and getting an ”is this a sausage?”-image interpreter online in a very short time. It can also require a PhD, development of a new programming language and perhaps inventing entirely new hardware components.
And the same goes for testing.