Last Thursday, June 13, 2019, we opened our new office to our clients and associates for the first time and at the same time celebrated the fact that we came together for the 30th time to share our knowledge and experiences in the form of a Dialogue Thursday (DiDo).

This time the theme of the DiDo was ‘From Hypothesis to Design. Conversion optimization is a multidisciplinary business in which data analysts, psychologists and UX designers join forces to improve the online dialogue between organizations and their customers.

An important step in this exchange is the iterative interaction between the psychologist and the UX designer. After all, how do you translate a hypothesis into a valid test design? How do you test in the simplest (or, in other words, cheapest) way possible whether your hypothesis can be confirmed or rejected? And does the new design actually cause a change in online behavior?

Jorrin Quest and Rowan van Sonsbeek shared their experiences as UX designers and offered the audience concrete tools to get started themselves at the end of the day during the workshop.

Jorrin Quest | Senior UX designer @ Online Dialogue

96% of all American households always have a bottle of ketchup in the refrigerator. In particular, Heinz Ketchup is a very popular and distinctive product, but also one that has gone through different shapes and designs over the years.

The first variants of the ketchup bottle, as I am sure many remember, were made of glass. A familiar design, not only for its appearance, but especially for the slightly aggressive movements required to get some ketchup out of the bottle.

The struggles experienced by consumers were recognized by the company, resulting in a new design: a plastic squeeze bottle. Certainly a step in the right direction, but again Heinz saw something happen that they didn't anticipate. People were turning the bottle upside down to make it even easier to get the ketchup out of the bottle.

This was another reason for a new design. The bottle was made shorter and wider so it could easily stand upside down, and the label was flipped.

Start from the user experience

The example of Heinz ketchup shows very well how to adapt a product step by step based on the user experience. Therefore, according to Jorrin, this is exactly what UX (User Experience) stands for: “the experience through interaction with an object in a context.”.

The design where the bottle stands upside down was not conceived because it was a good idea or a spontaneous hunch. If you had asked designers how they would design the perfect ketchup bottle, they probably would not have said they wanted a bottle that stands upside down.

Shorten your time in ideation and design

When we understand the user experience, we can write a hypothesis. In the optimization process at Online Dialogue, the hypothesis is written by the psychologist. Then, as a UX'er, you start working with this hypothesis and you want to come up with different ideas as soon as possible to test the hypothesis. When you have a number of ideas in place, you choose the best option and you start working that into a design.

While conceiving and designing, keep in mind that only 1 in 4 experiments produces a significant positive result. In other words, 75% of your design time yields nothing. Therefore, be critical of the time you invest per A/B test. This way, you can very quickly deliver more value by running more experiments faster.

UX for CRO checklist

What exactly should you pay attention to when creating a design for an A/B test? Which design do you choose from the many ideas you have come up with? Jorrin ended his presentation with some handy checks that are going to help with this choice:

• Does design cause a change in behavior? Often we see that designers or UX'ers are too preoccupied with visual changes, forgetting about behavior. But it is precisely the behavior of the visitor that you want to change, not the look of the platform.
• Is it technically feasible? It saves tremendously in time and money when you strive to test the hypothesis as simply as possible. And by simple, we mean as few development hours as possible. Test a simplified design first to find out what the effect of your hypothesis is before you put all that effort into a major modification.
• Is the hypothesis being tested? Sometimes we feel the need to adjust an extra element on the page because it looks nicer or maybe because you found a mistake or bug somewhere. All these adjustments may or may not have an effect on your results. Make sure your design is clean and really only one factor differs from the control variant. Only then can you say with reasonable certainty what caused the change in your visitors' behavior.

Rowan van Sonsbeek | Senior UX designer @ Swiss Sense

When you enter an organization as a UX'er, often very few people know how best to use him or her. For example, Rowan regularly received design assignments that did not seem to be a solution at all to the problem at hand in the user experience. In many organizations, people tend to think and act based on what competitors are doing and what our gut feeling tells us, rather than actually researching the root cause of the problem.

But as a UXer, according to Rowan, you're not there to just make something pretty. You look for the cause of the problem in the visitor's user experience and try to figure out how best to solve the problem.

Problem and solution: the UX designer paradox

Rowan shared an example from a supermarket where she used to work: after analyzing the behavior on the site, they saw that visitors were working very hard to put all their products in their basket but then still did not checkout. According to them, the solution lay in building a mobile application, as Jumbo and Albert Heijn have as well.

But when they actually started looking at the user experience, they saw that the problem was caused somewhere else entirely. Because supermarket products are delivered not from a distribution center but from a nearby store, visitors were asked for their zip code after choosing their products. As it turned out, not all supermarkets could deliver all products. As a result, visitors left the website without paying.

UX for CRO

Many people think they know exactly what is going to improve the user experience of visitors on a site. Unfortunately, in a lot of cases, this gut feeling is incorrect. And that's exactly why experimentation is so important. It shows you what works or doesn't work for the visitor and thus where your gut feeling was wrong. Working this way also gives you more and more support as a UX'er within an organization. You can show with hard evidence how a particular user experience can be improved.

You are a UX manager

The UX'er often finds himself between many different functions and departments. After all, you need a lot of input to really understand the visitor's problem, and in this you are often dependent on others. Make sure that you gather the right people around you. If you are missing certain information or a certain discipline in your team, be creative. For example, do your own research and ask some questions at customer service or visit the store itself.

Experimentation is the key to success

Celebrate all your experiments, whether a loser or a winner. Both will make you smarter, so both should be celebrated. And the more you learn, the better you'll be able to optimize the visitor's user experience. Just remember that one success is not a success story for everything. The optimization process is an ongoing one. Things change, and all these changes affect the user experience. Make sure you keep testing small changes continuously and stay on top of the user experience.

Workshop: NS International

After Jorrin and Rowan shared their knowledge and experiences, it was time for the visitors to go through the process from hypothesis to design themselves. Eline van Baal (Online Dialogue) and Thijs Nieuwdorp (NS International) shared important information about visitor behavior at NS International after which everyone got to work in groups.

All groups were given a web page as a case, information about the behavior on that page and a hypothesis. The goal was to work out a test design based on the information given.

Cooperation and exchange for quality

By experiencing the process step by step yourself and thinking carefully about the different phases you go through, the many options available to you and choices you make, you will gain a better understanding of the role of the UX designer, but especially how this discipline in turn is closely linked to the other disciplines within the CRO process.

Without this collaboration, mutual exchange and iterations between the various disciplines, establishing a successful CRO program will be difficult or even impossible. All disciplines depend on each other in achieving their individual and joint goals. Thus, the quality of one discipline's work affects the quality of the rest the team.

Want to know more about the 30th Dialogue Thursday? Watch the above compilation video for an overview of the afternoon. Sign up for our newsletter If you want to keep up to date with all our events! The next Dialogue Thursday is already scheduled for Oct. 31, 2019.

Photography by Susanne Ottenheym