This episode was recorded LIVE at the EuroSTAR Conference in Stockholm.
Thank you to our Community Hosts for Season One, Russell Craxford from the UK Department of Work and Pensions, and Gek Yeo of AIPCA.
On this episode, Gek and Willem Keesman (ENFP test leader at QNOORD, who was a finalist for the Rising Star Award 2024 with his community approach with Testpedition and Deep Dive Speaker at Automation Star 2024) talk about building chatbots, connecting with other testers, and building strong communities. We hope you enjoy!
Gek Yeo: Hello, everyone. Welcome to Eurostar podcast studio. I’m Gek, the lead manager for QA and release with AIPCA CIMA.
Willem Keesman: Hello, Gek.
Gek Yeo: And today I’m very lucky to have…
Willem Keesman: Willem. I’m from the Netherlands. , I’ve been in software testing for, roughly over 10 years now. I have a business background. I started it with a small business and a retail management study. And then I just gradually moved into more technical roles and ended up in software testing.
So from customer service, to content management and, technical support, and then eventually started testing and, Yeah, I’ve been working for a large energy company that’s called NUON in the Netherlands. It’s now known as Vattenfall. It’s, I think now it’s from Sweden. And worked for some startups in the financial sector, for the government.
And actually last week I started at Ordina, which is a daughter company of, Soprasteria, which is, a large company.
So this is my second week at my new job and I right away get to go to Eurostar, which is awesome.
Gek Yeo: He is very lucky, very lucky. And so, uh, what is the first testing project you have done then?
Willem Keesman: My first testing project was testing the mobile applications for the energy company, both on Android and iOS, but we didn’t automate it yet. So we had a whole briefcase filled with all sorts of devices, Samsung, apples, and we had to test it on the actual devices. So that was a lot of fun. Yeah. I remember.
And I had to travel. Yeah. I travel by train,, and could work in the train. So I had my suitcase and all these devices and I could work for two hours just deploying the app and testing the new devices.
Gek Yeo: Yeah. I always remember when we have like, um, not devices in the cloud, but rather physical devices.
Yeah. And you just have to keep. Track of all the old devices you have and then keep buying new ones. So that was quite fun. Yeah, and
Willem Keesman: also the the version numbers were also really important.
Gek Yeo: So true.
Willem Keesman:Yeah, I want to go back to a certain Android version. I think it was still called like Ice cream sandwich and lollipop or something back in the days.
Gek Yeo:Yes.
Willem Keesman:And yeah, that was a lot of fun. Different devices, different versions, so. Yeah. So
Gek Yeo: from your first mobile app testing projects, you have moved on. To different projects.
Willem Keesman: Yes.
Gek Yeo: So how do you see the change then in the trend for you?
Willem Keesman: For me, I shifted more towards automation. Because , I don’t like to do things multiple times over and over manually.
So to make my work more interesting, I just automate the more boring part and, focus more time on actual testing. Because if I say manual testing and Michael Bolton listens to this podcast and, uh, he will shout at me. No, that’s for me, that’s just the actual testing and automation is just a supporting activity.
Gek Yeo: I like automation testing. We get to save a lot of time and avoid, repetitive actions. And, when you work with different tester in, in environment, right? How do you encourage someone who’s, you know, start off traditional manual testing into automation? Because you follow the same path, right?
Willem Keesman: Yeah.
Gek Yeo: Manual testing on the devices and then now you are passionate in automation testing. Yeah. So how will you encourage someone who, because there are people who are still in that manual testing.
Willem Keesman: Yeah, I know actually. And I, a few weeks ago, I actually spoke to someone who just recently switched from testing in Excel sheets with all the scenarios to using a tool.
And that’s Tosca Commander that they use. And I was like, so surprised. I didn’t hear it for a long time. Yeah, yeah. There’s still people testing in Excel sheets and,
Gek Yeo: Putting screenshots in Excel sheets.
Willem Keesman: sheets and the green check marks behind each test case., how would I encourage them?
If they’re happy and good at what they’re doing, I think by all means do it that way. If you’re, if you Can enjoy yourself and are a good tester in that way, but stay on that path. But if you think , Oh, this job is really boring. I have to do the same thing over and over again. Then I would just suggest to start with a tool and then maybe give them a short demo.
Gek Yeo: Let them play around. But, In no means force it. So not like saying, Oh, we’re doing this, or everyone must do, must use this tool because that works well, counterproductive actually. True. Yeah.
Gek Yeo: You want to show them how you can use the tools to automate the process, make them enjoy testing. And then they will say, Oh, look, I saved time by using JIRA now, not the spreadsheets. And I can, you know,
Willem Keesman: Yeah.
Gek Yeo: Right. My test cases and things I keep record quicker. I think,
Willem Keesman: yeah, you have to think also what’s in it for them. Why would they switch? And if they have been doing things for a lot of years or the same way, well, that could be good reasoning behind that. But yeah, it could also just be them not knowing what else is out there. And just talking, sharing within a testing community, that helps a lot. So there’s maybe other teams that are more advanced in automation. They can give demos and what it brought them, the lessons they learned. And yeah, some people might join that, team as well or learn from them.
Gek Yeo: True. I think it’s also important to give them confidence, right?
Yeah. Sometimes it’s the fear, right? They say, if you are in a big group, everybody’s talk about automations. And if you are the only one who’s still doing manual, you feel very scared. If you’re like, Oh, okay. Look at them. It’s like, okay.
Willem Keesman: Yeah, I was just in a talk from Linda about the imposter syndrome.
And, that’s. Also how I felt in the beginning because I don’t have a technical background. So when I went first into automation, like everything was new, all the concepts, all the code things and the frameworks. So I just had to really write everything down every time. Sometimes I would just sit at a meeting just making notes and writing concepts down, googling them afterwards, doing e learnings just to catch up.
You know, there were lots of people with an informatics or computer science background. And I just had a business background. So it was all new for me. So I can really understand for testers that, have, yeah, are used to just testing manually or just used to testing with Excel sheets or for example, a full exploratory without, any charter backing it.
I think for them, it’s quite a big step. So you have to really be careful how you bring it and, yeah. What do you expect also from them? Because especially if you move from. Tools to open source frameworks. That’s even if you know how to automate in a tool, like, Tosca commander or postman or something else, and then you have to move actually to the I.E. That’s even a bigger step. That’s that’s really intimidating.
Gek Yeo: Yeah, especially from postman. Um, well, you go to, writing code space like rest assured you want to do karate and A. P. I. Testing. Yeah, that is Yes. But I find it quite interesting to move across and get to learn different kinds.
Willem Keesman: Mm-hmm .
Gek Yeo: And so what is the most, out of all these projects, right? Any memorable, interesting, you know, automation project you have?
Willem Keesman: I think I had the most fun that , was testing a chatbot. There wasn’t , a chat bot at a FinTech startup where I worked
Gek Yeo: Uhhuh chat bot. Yeah.
Willem Keesman: The company was called Afterpay. But it’s now part of Arvato Financial Services and they launched the chatbot and it was just really fun to bypass this chatbot and, try to get it to give strange answers and see how it responds to weird inputs. And at a certain point we noticed that we could inject scripts into the chatbot. So we took the whole JetBot code and put it into the JetBot.
So there was an embedded JetBot in the JetBot. So that was really a lot of fun. Yeah, the developers didn’t like that, but we had a, blast, of course.
Gek Yeo: Yeah, yeah, to fight all the bugs, right, coming out of it. Yeah, yeah, I can imagine that.
Willem Keesman: That was, uh, that was good times. That was good times, yeah.
Gek Yeo: I think that’s the thing about testing.
Willem Keesman:
And for you, was there also a moment for you that you were like, oh, this is really
Gek Yeo: Oh, I did, I did. What was really fun for me was at the beginning that we realized that we can actually, automate the checking emails So in the past we used to be have to go into individual mailbox and then check it and then click the sign up Yeah, once we realized we can automate it.
That was really fun because no longer we need to go in. Oh, that’s great. Yeah Check. Okay, all test pass is gone. Yeah, it makes us feel so much happier
Willem Keesman: Yep I can understand. Yeah, if you have to go to a mailbox each time, click a button, that’s,
Gek Yeo: oh, yeah, to check the verification link work and things like that.
And that was like, yeah, that was one of the more fun project that I got. And other than that, I think throughout the years, I realized that, slowly, we have talking more and more about AI. Whether AI is being used in the product that we use to test, or we use AI to do testing, and what is your, feeling about it then?
Willem Keesman: I think it’s been around already for a really long time, because I think the most, insurance companies and other large companies already use these large language models and algorithms behind the scenes. , AI is really popular. I use it mostly as a sparring partner, so I just have to do exchange IDs and I had also a discussion with some people about this the other day.
I think you should not blindly rely on it. And you have to know what the output means. I mean, if you just say, generate me a framework, rest assured, and give me the classes, give me the POM if you’re in Maven. You put everything in the idea and it probably will work, but then one day it won’t work and what do you do then?
And then you again have to go to share TPT with all the errors and ask, Oh, this is happening, but you don’t really know what’s happening behind the scenes. Yes, you
Gek Yeo: need to have to be able to establish the trust in the AI to in order to give testing over. Yeah. So in this case, as you mentioned, how do you manage to keep contact?
Willem Keesman: And with the committee in Netherlands, we got about testing in the Netherlands, yeah, and I invented a concept in 2023. It’s called TESpedition. Okay. And I’m actually also partaking for the rising star here in the Eurostar. Okay. So it’s in the final. So fingers crossed, we will know, in the late afternoon.
Okay. Alright. And a test expedit edition is basically a concept where, with a group of testers, you visit a company, a company opens its doors and, shares, its testing challenges. And you basically have a whole day, of talking about testing with experts and about the real story. So not the sugarcoated, marketing stories like, oh, we use the greatest automation tools, we have the best modern framework.
No, it’s. Okay. , we have really problems with production sensitive data. , we have lots of test environments and teams are interfering with each other. , we have a fragmentation of tooling and best practices. These are the topics we address during such a expedition. And we also keep in touch later on.
We share the lessons learned. So once such a expedition is over, it’s not like, Oh, we visit nice. We shake hands. We never see each other again. There’s Contact on a structural base. And that’s really great.
Gek Yeo: Okay. So how do you manage to, establish this contact structure base? Then how do you make sure that the contact is continuing and not die off?
Willem Keesman: Yes. Like everything. Uh, right now it’s just by, by some shared task boards and things like that on the internet, but I want to really make this platform like this test, but a testpedition. com platform where you can just, host a testpedition, join a testpedition and. have an archive that’s freely available for everyone just to read what lessons were learned from company visits.
And yeah, it also depends how open and transparent a company can be. Because for example, we had a keynote yesterday about the SAP Gripen fighter. Well, I don’t think they want to put all the lessons learned in the internet. Yes, I think some other parties would be interested, but that would not be a good idea. But in general, I think it’s really nice , if it just doesn’t end there if you’re just not Visit, but you keep in touch. You keep interacting with each other. And that’s yeah.
Gek Yeo: Yeah, keep sharing ideas. Yeah. Yeah, I like about it Yeah, we have like people that we met over like Conferences and after that we still stay in touch. We share blog posts. We also You know, listen to the podcast. Yeah, I think that is a great way.
Willem Keesman: But also here, what are you doing here with Eurostar Huddle? I mean, you have this interesting seminars and these lunch sessions. There could be more, in my opinion, because we had a nice one, I think, with Gaspard one time and just also exchanging ideas on AI. And these are the type of things that you really get to connect with. Community engagement going and yeah, because
Gek Yeo: You get to hear different company culture. Yeah, everybody is different, that’s and
Willem Keesman: but the different but we also face a lot of the same problems. That’s true. Yeah, some issues you hear everywhere, whether you go to a government company or start up or well established name in the financial services, some challenges are everywhere.
Gek Yeo: Yes, yeah, test data.
Willem Keesman: Yeah, test data. Yeah, yeah. To name one. Yeah,
Gek Yeo: yes. So, um, how would you think, of testing be like for you five years down the road?
Willem Keesman: That’s a difficult one to answer because I would not expect that I would be doing what I’m doing now five years ago because I was really more really busy with the tools and the tech and less with the people in the community. For me, I really want to move more into a role where I’m just. Have a lot of, diversity interaction with different people,
Gek Yeo: Coaching them…
Willem Keesman: giving workshops. I really love workshops. The format, because when I’m hosting a workshop somewhere, I also learn a lot myself because I get lots of feedback from the crowd.
But I also see I just sometimes just sit at a table. I say nothing. I just listen to the conversations going on, and I hear really great things. Yeah, and it doesn’t, it’s not only coming from the. The seasoned testing professionals like the gurus, but sometimes there’s a junior there just started and has just like, Whoa, I never thought of it this way, you know, like this.Yeah. I’ve been doing this like this for 10 years. But okay, what you’re saying also really makes sense. So yeah.
Gek Yeo: That’s that’s quite interesting. Yeah, because when I first start off also, When we were doing testing, we didn’t really have this like test driven development where, you know, you write the test at the beginning when you write the code, test case will normally written towards the end.
Willem Keesman: Yes.
Gek Yeo: Yeah. So it’s not changing things. So
Willem Keesman: with that, but I think that these kind of developments are really good. What you mentioned test driven because then you start thinking about product quality at a very early stage. Yes. And if this. Tests that are written are based on the risks. Then it’s even better because everything is Related to risks.
So no risk, no test that. Okay. I heard that somewhere that I, that, that makes sense. I think,
Gek Yeo: yeah. Testing is risk. Risk management. Yeah. Yeah. Was because they saw,
Willem Keesman: I saw that yesterday also in the Swiss cheese in the presentation. The first keynote, I think it was, yeah. That was good. And I saw also risk, and on the right side, I thought, yeah.
It’s all starts with risk. Yes. Without the risk. Yeah, why would you just make all these tests?
Gek Yeo: Yes, these are great with you too. So, is there anything you would like to give to our audience before we close?
Willem Keesman: Well, I’m Always open for a chat with anyone. So I, if you want to connect with me on LinkedIn or you see me at Eurostar at another conference, just, come over. Let’s have a chat about testing. I love the profession. I love the community and,, I’m looking forward to hearing from everyone. And I would, also like to thank you for, giving me the opportunity to, speak.
Gek Yeo: You’re most welcome. I’m very happy to know you. Likewise. Yeah. Thank you everyone.
And that’s the end of our podcast. Thank you.
Willem Keesman
ENFP Test leader at QNOORD
Willem hosted some workshops and internal events for smaller groups thus far and I am very excited to take it to the next level. EuroSTAR and AutomationSTAR were the highlights of 2023 for me, experienced a lot of positive energy and made new friends in the testing community. Being able to actively contribute to such events would be a privilege. Apart from testing I like to ride rollercoasters, watch horror films and play RPG games. Every once in a while I escape to remote locations and go for long walks in order to