Why a portfolio piece on a simple button?

Well, this project wasn’t as simple as it seemed. This was a result of my work, and I am pretty proud of that one. To make the case more understandable, I need you to go back a few years. When I joined Briq, I knew that you must create a UX foundation as the first UX professional. After the first stakeholder interviews with colleagues from different departments, I documented some challenges I needed to work on. One of them was the validation of ideas and completed work.

How it was…

The company operated by getting features from clients, developing them, and releasing them. After the release, they wait if the customer service gets complaints about the delivered work. Not a bad thing for a start-up when you need to pay attention to multiple items. But this is something I can help with.

Screenshot of a Google Analytics dashboard.

Validation

As you all know, we can validate work in two ways: qualitative and quantitative research. In qualitative research, you, for example, do interviews or user testing. In qualitative research, you often look at data in some way. Combining these in our work is always better, but we need to start somewhere. In the beginning, I wanted to do some user testing to see the status of the applications. Because Briq had no experience in this, I created a proposal of how this would look and calculated the time and price I needed to do this. Maybe I was a little bit too honest, but they were scared to invest money and time in something they didn’t know the result. Fair enough. I realized I needed to make it smaller and more straightforward. After talking to one marketing colleague, I heard he was working on implementing Google Analytics into the marketing website. I realized that this was something easy to implement in the applications. Because they were web-based. After some consideration, they decided to implement this into the Funnel. By showing the data and insights we could get from this, the product team became a little bit more excited. This created a way also to implement Google Optimise, the tool where you can do A/B testing. This will be an essential tool for the upcoming years of Briq.

Finding a use case

To show the value of the tool, I needed a small project. When I looked at the Google Analytics data, I noticed that customers on mobile stay relatively long, compared to desktop, on the lists page before they continue into the funnel. From my previous experience at bol.com, I assumed this could be because of the number of deals on the lists page. I know deciding which deal to pick becomes more challenging when the number of options becomes larger and larger. I couldn’t simply remove deals, but I could do something else, highlight the filter button. Customers can use the filters to decrease the number of options based on their input. That was something I wanted to do.

Screenshot of a Google Optimize results page.

But how do I do this…?

It was a great feeling to have a small project to show the value of this tool and the process. But how do I create A/B tests and interpret those? So I need to read up a little bit. After reading Medium articles about A/B testing, the foundation of what it is, how to use it, what significance means, and what the Bayesian and Z test calculation means. I am more comfortable doing it. It also helped a lot that I found tools to help me in the pre-analysis and post-analysis. So now, I was able to create the test. It was simply adding a background color to the filter button. Because I have CSS knowledge and already have a selection of colors to choose from, it was an easy set-up. After informing the company about the idea and the test. I was able to test this for one client, and the first A/B test I made went live. It was a fantastic feeling!

The slides I used to show the results of the A/B test

How did it go?

It went pretty great. After two weeks, I was able to analyze the results. Fortunately, these were positive! Because I needed to sell this, I needed an excellent presentation. So created a slide deck in Figma where I can make the data shine more than it could in Powerpoint or something else. After that, I created a great message in Slack about the learnings and invited colleagues to comment on it. It was a great way to make people enthusiastic about this form of validation.

Learnings

This project was interesting from the perspective of how to steer the company in a direction you know is right. Ideally, you want to have all the validation methods in your toolkit from the start, but taking little steps towards it helped to create a business case for other things, like introducing Hotjar to getting hired by clients to do user research and eventually have a discovery team to have more time and resources for research topics. It feels great to look back at this and see your impact. The most significant learning I got was a better understanding of the company's status on a specific topic and adjusting your tactic to it.

Curious for more content?

Project: Redesign Planboard

Project: Weekschema