Karl Gilis
Co-founder
AGConsult

Karl is co-founder of AGConsult, a Belgium-based usability and conversion optimization company. He is number 3 on the list of most influential conversion experts in the world and will be holding a session and a workshop at the Conversion Conference in Berlin in November 2016.

Karl, have you ever been to Berlin? What are the things or places you are looking forward to the most regarding your trip to Germany?

Yes, I was there last year also as a speaker at Conversion Conference and I enjoyed it a lot. It was my first time in Berlin, so I walked around at “Unter den Linden” and ”Checkpoint Charlie”. Seeing the “Reichstag” was very impressive. The thing I liked most during that visit was seeing the “Pegamonmuseum” because I studied art history at university.

I am really looking forward to visiting Berlin again because there are a lot of things I didn’t do last time: I want to visit the “dome of the Reichstag”. And of course there are so many museums I haven’t been to and “the Memorial to the Murdered Jews of Europe”.

Another reason to go back to Berlin is the excellent food: I had my birthday dinner at the “Reinstoff” and that was one of the best dinners I’ve ever had – so I will definitely go there again this year.

You are working in this business for more than 15 years now. How did website optimization change during this time?

I think in 2001 the words “conversion optimization” didn’t exist yet. That’s why I still call myself a “usability expert and information architect” because in those days it was all about “usability”.

Although we didn’t have Google Analytics in 2001, there were some other analytics tools to analyze websites. We also did a lot of user testing. That was our basic method: seeing real users interacting with a website. We also used surveys a lot. The issue was that there weren’t any good AB-testing tools at the time. So we had to come up with our ideas based on the user testing, analytics and surveys, implement them and then hopefully we would see in the next user test or in the analytics data that what we changed on the website made a difference.

In the 15 years I’ve been in this field, and especially in the last 5 to 7 years, there has been a rise in good and affordable tools like Google Analytics, Kissmetrics, Heap Analytics and Hotjar. I really love that there are so many tools available now that help give us better insights.

And of course the AB-testing tools like Optimizely and Visual Web Optimizer (VWO) where you can test if your ideas and the solutions you came up with will actually make a difference.

The title of your session sounds very interesting: „Don’t Do Anything On Your Website You Wouldn’t Do On A First Date“. Can you please tell us how you came up with the idea for this title?

I always want to refer to real life and tell a story in my session, because I noticed that it is easier for people to remember the facts if they have a story to remember: So my story is about “first dates” because everybody has been on a first date and has made mistakes on that first date.
Some time ago I started to see similarities between the basic dating tips and websites. A typical dating advice is like “Make sure you show up on time: don’t keep your date waiting”. And the same goes for websites: Users don’t like to wait, they like to get results immediately when they click on something.

Now we are curious about your first date – what wouldn’t you do again?

My first date was about 30 years ago, with the girl who is now my wife. Given that we’re still married 30 years later, I think our first date went fine. I would do everything the same – maybe drink a little less. ☺

And when I transfer this to websites my general advice is: don’t be drunk when you make your website. Just kidding. I don’t think many people are drunk when they’re making a website, but with some sites it’s hard to tell.

In your opinion, what are examples of good websites, websites that convert well?

That is a difficult question and to be honest I don’t really like it because often people start to copy these best practice examples and that’s not what it’s about. There are lot of good websites out there…

The thing is not to copy what others are doing but to find out what the problems are on your site and to fix those problems. Through research and testing. Boring, but true.

Your abstract says you are “obsessed with AB-testing“ – please tell us more about this and why everybody should be AB-testing?

Obsessed is maybe a bit exaggerated but I really, really love AB-testing. In the past we only saw after we actually implemented an idea if it increased conversions. AB-testing gives us the possibility to test the ideas – always based on user research not on gut feeling – before the final implementation and see if it will work or not.

AB-testing also allows you to test small changes as well as huge changes. Additionally you can do follow–up AB-tests for your improvements so you can keep refining your original idea and make it even better.

A lot of people are still randomly changing things on their website based on gut feeling. What THEY like and think is beautiful and not what their customers like – so I hope in the end AB-testing will be the big winner.

You have done a lot of user tests. What is the most common mistake that people make in this process?

User testing is not that difficult but there are a lot of mistakes people make. One of the biggest mistakes for me is, that they work out a list with questions they want to ask and then literally read them out. If you want to observe natural user behavior your interaction with the user should be natural. It should be more of a conversation than a test.

Of course you need a list of questions, but you should ask them in a natural way and adapt the question to the user. One of the biggest mistakes in user testing is asking the wrong questions that lead the test person too much in a certain direction.

You will be holding a workshop in Berlin as well. What can participants expect from this workshop?

I’ll be doing the workshop together with Els Aerts, my business partner in AGConsult for the past 15 years. She has even more experience in user testing than I do. So I thought it would be a good idea to do it together.

We’ll talk about how to do user research, especially user testing in the best possible way. That will involve everything concerning moderated user tests, for example:

• How many participants do you need?
• What is important when you select the participants?
• Recruiting special testing groups like children or older people.
• Crafting the test scenario.
• How to write the questions down, how to ask the questions and adapt them to individual users
• Questions about user testing on competitor websites
• Use of different tools
• And much more…

At the end of the workshop the participants will be ready to do user-tests on their own.

Please tell us about your typical day in the office.

I am traveling a lot and giving workshops or trainings so I don’t have a lot of typical days in the office.

What are the most important things on your office desk?

I am a very chaotic worker – so I have chaos everywhere on my desk. My laptop is probably the most important thing, I use it all the time and it is my brain. Just as important as the laptop is paper and pen, to write ideas down.