In this blog series, we feature testing and quality professionals from various industries. The Testersuite Team likes to hear from these professionals the various views on testing and software quality. In this edition of Let's Talk About Test you will meet Fokko Keuning, unit manager Qualitiy Solutions at myBrand Conclusion.
"Understanding quality is important to identify and mitigate risks earlier."
Just a heads up: who is Fokko?
Of course, I am not just a unit manager of Quality Solutions at myBrand Conclusion. Meanwhile, I am 50 years old, live with my girlfriend, have two daughters and two bonus children. We live in Dronten.
Since August we have had a Tervueren shepherd puppy alongside our retired guide dog. Maybe the puppy can learn something from the old dog.
In elementary school, did you want to be a quality manager?
No, I had only one passion and that was to become a pilot. I did everything for that. In the Air Force I was one of the few who passed all the examinations without any problems. I realized that becoming an F16 pilot was not going to work out. As a child I had a mild form of epilepsy, which I got over at the age of 12.
Becoming a helicopter pilot was my aspiration, and they saw no problem with that in the Air Force either. I even passed the final simulator tests without a problem. In the end, the medical staff rejected me. I was declared 100% healthy but the Air Force did not want to go along with that.
Aviation is still my quiet passion. At home on the simulator, I fly an Airbus A320 and it goes great.
My second passion is IT. As a 12-year-old kid, I had a computer without a hard drive yet. You know, with floppy disks for the a- and b-disk. Later, I had a personal computer. In those days that was called IBM-compatible. In college, I still tried the direction of HR but went in the direction of IT.
So how did you get into the world of quality management?
For me it is important that something looks good and is good. In IT, I moved into project management through roles such as network administrator and IT management. At one point, the number of assignments at the organization where I was working at the time ran out. As a result, I had to do something else.
Then I became a driving instructor; IT was ready for me for a while. So I started my own driving school in Lelystad and a franchise driving school in Zwolle. The idea was to get out of the car myself and become a full-time entrepreneur.
Meanwhile, my first child had been born and then you miss a lot as an entrepreneur. This prompted me to look for a job in IT again. There was a vacancy as a QA consultant and I thought it would be great. Because of the gap in my CV, I was hired at a somewhat lower level. However, after three months that was quickly changed. That took place in 2006. Now I end up being unit manager Quality Solutions at myBrand Conclusion.
"Start as early as possible to take out mistakes and reduce risks."
How important is testing within quality management?
Nowadays, the shift-left principle is increasingly embraced. In other words, start testing earlier. This is an important part of Quality Assurance. Start as early as possible to get bugs out and reduce risks. In doing so, you also reduce recovery costs.
Worldwide, 81% of IT projects are still delivered late. As many as 51% still have issues in the production environment and 41% go over budget. When I compare that to figures I have been communicating in training sessions for years, we are still not doing well in IT. Testing is so important but it remains an underserved child.
Why is there still no proper testing in IT?
Last year I gave a presentation to the management of a client in which I had to explain why they still had to perform acceptance tests themselves. They didn't understand why that was necessary, because surely myBrand Conclusion had already tested?
I explained that we did indeed test. But ultimately it is your business, it is your business processes and systems. You yourself can best determine whether everything works as it should. You always see it happen when the key-users start testing that the defects then flows in. This is because the key-users work very differently with the systems than we do as outsiders.
So the myBrand Conclusion consultants really need to bring clients into this?
Yes indeed. For example, I always explain that in addition to risk-based testing, you should also pay attention to value-based testing. Risk-based testing is based on the no-risk-no-test principle. With value-based testing, you look at where the most value lies for the customer.
Let me cite an example: A customer brought products to the cloud. We then naturally start testing functionality. However, the real value is in performance and security. This is because they are going to the cloud. So that was the primary focus of our testing.
That's the additional question you have to ask as a test consultant, where is the value to the customer. In a previous Let's talk about test, Dirk Janssen mentioned that it's about processes and risk. I complement this with where is the value to the customer. Agile and DevOps are also based on value so that fits in perfectly.
"For SaaS applications, you also need to test."
Should you also test with Low-code and SaaS?
My former colleague Jan-Jaap Cannegieter had made a presentation on the question of whether you should also test in low-code. Yes, you do! Low-code works with standard platforms, but actually you develop custom applications. If anything needs testing, it's customization.
For SaaS applications, you also have to test. When I look at the new public cloud propositions from SAP, they offer them based on best practices. As a SaaS user you have no influence on the platforms, that lies with SAP.
The question is whether your business processes are supported the way you want. There's customization in that, too, which we often fill in with low-code solutions. So you have to test. The most important question is whether your business processes are supported in the right way. Regardless of whether it's on-premise or in the cloud.
So testing remains important!
myBrand Conclusion uses Gartner's pace-layered model. The first layer is the system of record. That's your core legacy system. You want to change that as little as possible. Then you have the system of differentiation. That contains applications that enable an organization's unique capabilities. The third layer is the system of innovation. That's where we position the low-code platforms. The various systems communicate with each other, and the last thing you want is for bugs to appear between those layers. Hence, testing remains important.
Where are your challenges with customers?
Unfortunately, this is mostly on raising awareness and convincing customers to invest in quality and testing. That takes a lot of presales time. Once the consultants are in, it goes well and people quickly see the added value. Then I think, yes, I've made someone aware again. My goal is to make people quality conscious.
Why is quality so important?
Understanding quality is important to identify and mitigate risks earlier. This prevents rework and saves money. The impression is given that quality costs money but it actually saves money because you prevent rework.
Quality allows you to deliver faster, time-to-market is important, and quality is going to help you do that. If there is no quality and you are going to speed up, then that means mistakes and so that costs money.
"It often goes wrong at the requirements stage."
Boehm's law still remains, doesn't it?
It certainly stays that way. We once tried with a group of smart people to come up with a clever variant of Boehm's law(The cost of repair increases exponentially the later an error is found and repaired in the development process, ed.) That didn't work out. Because of its simplicity, this law still remains relevant today. The power of simplicity!
It often goes wrong at the requirements stage. Projects often step over these too quickly because they want to get started. An example: I have as a requirement that I want an off-road vehicle in which I can carry luggage. Then I get a Landrover in my driveway for testing, whereas I meant a golf cart. The conversion cost to turn the Landrover into a golf cart is huge.
You shouldn't think in solutions right away. It is better to ask questions first. That happens quite rarely these days. At myBrand Conclusion, we also help the customer with this.
Reviewing deliverables costs money they say. I guarantee an ROI on reviewing of at least 1:3. I even achieved an ROI of 1:20 with one client. In doing so, I avoided tons of repair costs for the client on large projects. It cost the client money to hire us but the savings it produced were many times greater.
How do you see the application Testersuite within your work
I was at an event the other day where I was asked what the most commonly used testing tool is. Everyone knows that this is Excel. I am striving to increase test maturity with clients and that's when you want to move away from Excel.
When I talk to customers they always say I have Excel and that's what we use. When I ask how much business risk you cover with that, they don't know. When I ask if there are production issues, the answer is 'yes'. If you then analyze, you see that you can cover a greater business risk with fewer test cases and therefore test better and cheaper.
In Testersuite you bring the right test cases properly into all your (sub)processes. It supports the entire testing process and once you have it set up properly, you can make it repeatable. Through the master list, you retrieve and reuse everything when needed. The workflow in Testersuite takes a piece of communication out of your hands for you. It offers you structure and a step up in test maturity. In addition, it is ideal in preparation for test-automation.
In a general sense, how do you see the future of the testing profession?
Very positive. I think we will always retain manual testing on the changes that are realized. There is a lot of thinking toward test-automation. That's what my unit focuses on as well. You also hear a lot about AI. Will everything soon be automated? How does AI interpret a requirement? No idea.
There are many wonderful developments, it is important to keep up. Quality assurance is very important if one is aiming for speed. People say that quality and speed go hand in hand. My opinion is that quality goes before speed!
What is your advice to other test professionals?
Always think from the value to the customer, that's what you do it for.
Anything else you want to say?
Make sure you do things you enjoy.
