Certification – is it worth it?

Another topic that has been on my mind is certification for testers.

It’s a tricky one, as there are different qualifications that a tester can get – there are the formal ISTQB or Certified Agile Tester (CAT) courses which will provide some sort of certification, specific qualifications that cover tools such as QTP, or a specialism of some sort such as Security testing.

A number of people have asked the same question, namely ‘is it worth getting a formal qualification?’.

My experience of taking ISEB (as it was then) after being a tester for 15 years was that I learned what I needed to pass the exam, ignoring what I do in my day job. I passed, and then ignored most of what we had learned as it wasn’t relevant when I went back to my day job!. The theory was great – but out of date – it hadn’t moved with the times.

There probably is value in testers taking the exam with some test experience, as opposed to just going straight in with no practical experience – you do need that to help you work out how to apply what you learn, but I do not see the relevance if you have been testing for a long time, as you have to ‘unlearn’ what you do, in order to learn what the examiner wants you to write down. That does not make any sense.

Unfortunately many employers are blind to this, and insist on a qualification, believing that it means that the person is a good tester – but it is not the case at all. I have interviewed people with ISTQB Foundation for example who have no idea how to approach testing when I give them a User Story as a practical test and ask them to outline their approach and give me some example tests. If they have a qualification (whatever it is), then surely they should know this!

The idea of having a standard industry certification is great, however unless that certification actually meets the needs of testers in the industry and reflects how organisations work, it has no practical value, and therein lies the problem. We are left in the position where employers (who often are not testers themselves) do not know how else to measure if a tester is any good or not so insist on what they perceive to be the standard.

We need a standard certification that is recognised and respected worldwide. We need it to be relevant to how the IT industry actually works. We need it to be kept up to date with new trends. We need it to be useful to employers.

We have an opportunity to make something of real practical value. That’s the challenge to the governing bodies, if they choose to accept it.


Making a meal of testing?

As an industry, are we making a meal of testing?

This might be a controversial posting, but I feel a debate is needed on this, and I am happy to start one!

Over the past 5 years, testing has changed, and so have the roles that testers and managers play in software development teams. As part of the change, I have noticed that as an industry we seem to be making things more complex than they necessarily need to be. There are so many courses, books, testing magazines, Linked-In and other blog posts (like this one!), meetups, etc etc, which are positive where they add value to how we do testing, but I can’t help feeling that there is a lot of job creation/justification going on. Much of the so-called ‘new thinking’ seems to be a rehash of older ideas, and then promoted by many people, each wanting to add their own stamp to things in order to stand out in the crowd. It creates a vast amount of noise.

Let’s go back the a basic fact – testing still has the same purpose as it always has done. It doesn’t matter how many different processes and models that people come up with – what we as testers are trying to do is to establish whether the code that has been written meets the end user’s requirements, and highlight where it doesn’t, logging defects found so they can be fixed. Okay, that’s very simplified, but isn’t that the point?

The Oxford dictionary definition of the verb ‘Testing’ is as follows: Take measures to check the quality, performance or reliability of (something), especially before putting it into widespread use or practice.

In order to do that we may need a number of things in place such as environments, data, user logins etc. We’ll need to consider what is in scope and what is not, consider what test types we need to do – functional tests, and non-functional tests – load, performance and security. We need to write the tests, the automation code, and once ready, run the tests, raise defects, retest and so on. Certain applications will require integration testing if there is a data flow across a number of applications, and for more complex systems, more complex tests must of course be written. We may have Apps which need to be tested on mobile simulators and devices which adds a new dimension to how we do testing, but no matter how simple or complex the system is, the aim remains the same.

The point here is that the overall testing process does not need to be rocket science, and it does not need to be over complicated with sets of process documentation that takes days to write, and complex models which no-one can use or understand.

Of course, this has a direct impact on those who are running courses on different complex test models, or who are offering consulting services. So my question is ‘Does this add value to our industry?’ I believe it is only a matter of time before those at the top who for so long have believed that testing actually is complex and needs consultants, training courses, a variety of models and approaches, wake up and start to question exactly what all this adds to the quality of their software for the price they are paying.

Agile has helped to break down into smaller chunks the code we are asked to deliver. We therefore can no longer hide behind processes and models – the team will quite rightly question why tasks against a user story need to be done at all if they are time consuming or appear to add little or no value.

Value for money is something we all have to provide nowadays. CTO’s are looking carefully at roles and what value they bring to an organisation. We run the risk of over egging the pudding here, and that will push back all the work that has been done to make Software Testing a career choice for people, rather than something to fall into as a last resort as a failed developer (not my words – this was said to me a number of times in the 1990’s!). Having a tester in a team is seen as positive where they add value – so let them add that value by producing good tests with the minimum amount of paperwork. Plan what is needed to do the job well by all means, but don’t over do it!

You may disagree with me, but I remain convinced that far too many people are making a meal out of testing, and it does us all a disservice.