During my 30+ years in IT, I have come across a torrent of misinformation and even deception regarding the benefits of implementing a Quality Assurance/Quality Control regime. Much of this misinformation has come from major companies who have gone on to benefit significantly from massaging the truth. I have even been asked to provide guidance that will increase business rather than assist the client – of course I didn’t do it, but they probably found another “sap” to do their bidding…
Often the problem stems from looking for a solution rather than analysing a problem. A simple directive (usually from a senior Exec) asking you to source a Test Automation tool (usually after they’ve been wined and dined by another senior Exec from a software vendor) can lead to all sorts of mayhem. Test Automation shelf-ware is a well-documented and often told story (there should be competition to identify the worst ever Testing Tool purchase), but how do you push back on the gung-ho Exec who wants to propel his/her solutions into the ether faster than you can boil an egg? The answer is quite easy – you do what all good Testers do and gather evidence. Not evidence regarding the solution (a myriad of test tools) but evidence to support the need (or not) that Test Automation is the answer to whatever the Exec considered was a problem.
If the problem is time to market, then buying a Test Tool is unlikely to address this in isolation. If the cost of specialist staff is the issue, then buying a Test Tool is also unlikely to address this in isolation. If the problem is….. you get my point, I’m sure; there are many solutions to many problems, you just have to make sure that you’re addressing the right problem before you go off hunting for the appropriate solution. A word of warning on this though, I have seen many Test Managers reduced to a cowering wreck because an Exec has already made up their mind, so make sure you have a Plan B and even a Plan C to head off the impending disaster at the pass – and “I told you so” isn’t a Plan B….
Back to lowering costs, speeding products to market, exceeding our customers expectations – “what if we hire a Consultant to improve our process”? Better still, let’s ask him/her to thoroughly review all of our Quality processes and then implement a best of breed solution. Execs love implementing best of breed or best practice solutions. It makes them feel warm and fuzzy when they produce a Quality Blueprint Report that states how all their Quality problems will be solved by implementing the recommendations in an Independent Report. After all, an independent Consultant told us!! Just for the record – my position on best practice/best of breed is that these concepts are just sales/marketing tools for making us feel comfortable about making difficult decisions/choices.
I’ve conducted many Process Reviews over the years and none of them have been successful because of the the glossy report we produced – they have been successful because an effective Change Management Program was initiated with specific goals and discrete outcomes, measured regularly and consistently for months (sometimes years) during and after the review. Businesses are always changing and if you assume that a Report written at a moment in time will solve all you problems you’re in for another disappointment – continuous monitoring and review is far more effective than consultant-led periodic assessments.
Implementing a Test Management tool does not give you control of your Testing activities. Implementing a (good) Test Management tool gives you a better chance at controlling your Testing initiatives but good process will always out-perform any tool or product. If you don’t have a triage process there is no way you will ever manage defects effectively. If you don’t link every test case back to a need or requirement you will never know whether or not your testing activity has achieved what it set out to (reduce the risk of failure and increase customer satisfaction).
Introducing a Test Centre of Excellence (TCoE) is not the answer to poor quality. Poor quality typically stems from poorly defined requirements, poor design or insufficient analysis – no amount of money spent on improving your Testing capability (if, in fact, a TCoE could guarantee this) will improve product quality if your fundamentals are broken. We all know that an ounce of prevention is worth a ton of rework and repair.
Reducing the amount of Testing you had planned to do, to meet a deadline, will not make your customers happy (if this increases the risk of failure). I have lost count of the number of times I have been asked to reduce the amount of Testing we had planned because earlier phases of the project have created time to market issues and “less Testing can help us make up our lost time”…. It’s even worse when we are asked to cull our test cases to “make the plan fit the schedule”. I ask you – what are we here for? If you are serious about quality (and my professional experiences to date lead me to think the majority of companies are not as serious as they would lead you to believe) then build a Quality Framework that questions the product at every stage of its life.
Investing in your staff is a great way to grow your capability but sending them all off to a Certification Program isn’t the answer. There are many less expensive ways to grow your capability and certification is only one way to improve capability. I see the main advantages to certification being the creation of a benchmark and the adherence to worldwide standards, there are far more effective ways to improve skills and ability. My belief is that you need to spend at least 2 weeks every year on the development of each and every one of your Test team. This doesn’t have to be conducted externally and doesn’t have to be expensive, but it does need to be done. I heard a very sobering statistic many years ago that 99% of organisations do not spend their Training and Development budget – usually due to the challenges of keeping the business running. I’ve also heard many times that when budget cuts are considered Training and Development is an easy target.
Dateline: August 6, 2013; Central Coast of NSW