At Coolblue, we want to do things every day a bit better. The same holds for our UX team, which grew from 7 to 14 members in the last eight months. We want to make our websites clear, useful and easy to use. But we also want people to experience pleasure and delight while using our websites. For this, it is essential to understand the behavior, context and needs of customers in more detail. And that is where UX research comes into play.
A lot has been written about UX research and how to do that optimally. However, there is no standard or best way of doing it. It depends on the product, the teams, the process, the organisation and so on. You need to find the way of testing that fits the organisation. In this post I want to share some background on how we do research at Coolblue and what we have learned.
The way of testing
One challenge we faced was how to integrate user tests into different phases of the design process. We know that testing earlier makes the design process more efficient. Also, it helps identifying main UX issues even before the development starts. On the other hand, this means also testing more often and asking extra time from team members to follow the sessions.
Our first step was to figure out the best frequence and way of testing that could fit within Coolblue. We started with trying two frequencies; every 2 or 4 weeks. Also we experimented with the number of participants; 3 – 6.
The main learnings were:
- Effort vesus gain. After the first few studies we learned that testing once a month with 5 participants fits best with our current process and teams. Testing every sprint is too frequent. We would have to invest a lot of time and effort without enough new insights.
- Sketching and prototyping. Introduction of more frequent tests stimulates the team to think more spontaneously about the user needs when designing. It also stimulates the teams to sketch and prototype new ideas more often.
- Iterative testing. Since we test more often, we can test prototypes iteratively. This way we can test whether the redesign solves the observed issues. And this was not always the case, illustrating the importance of iterative testing.
- Pragmatic testing. In addition to the regular UX studies we observed the need for more pragmatic, small tests. For example, tests with users in the shop or with some colleagues who are not directly involved in the project. Of course we do not use this kind of tests for making important decisions or for answering difficult questions. We are critical about the conclusions we can draw. However, these pragmatic tests are sufficient for gathering fast input for discussions in the team. Also, they help to see quickly whether an interaction works or to shed some new unbiased light on the project. Involving the user in the process does not have to be always expensive and time consuming. It all depends on your questions and the kind of data you want to gather.
The focus of UX research
With the regular monthly UX studies we are mainly focusing on validating our ideas and designs. The aim is to to optimize our work based on direct user input. It is very nice that we are doing that more often in the last months. However, this is not enough for understanding the whole spectrum of customer behavior and needs. Neither it is enough for identifying and understanding the main areas of improvement.
In addition to these regular tests, we are also focusing on:
- Personas and customer journey. In a recent project, we used insights from interviews with customers to create personas and to map the customer journey. These can help us to prioritize issues and to continuously relate designs to the overall context of user behaviour. Also, we will use them as a facilitator for involving different teams in the design process. This helps to create a common ground for starting a discussion or a new project. Feel free to share your experience with personas and customer journeys in your projects. We just started using them.
- Broad insight studies. Parallel to our running projects, we do a more broad insight study every 2 to 3 months. In this study we focus on a specific topic that will become relevant in the upcoming period. For example, in our last study we focussed on behaviour and needs of customers while shopping on mobile. Data from these studies give input for vision and strategy, helping us to match the customer needs with the business goals.
- Combining different sets of data. We experience that quantitative data usually are very convincing to stakeholders. However, sometimes the A/B tests or Google analytics data give strong quantitative insights but without a context. The advantage of working within a large organisation is that we have specialists in both the qualitative as quantitative domain. We don’t have to guess about the explanations. We can gain user insights during UX studies in order to find out the ‘why’ behind the observed quantitative data. The challenge for us is to combine these data more often.
UX research at Coolblue is just starting to feel as a part of the Coolblue way of working. We have sown the seeds and now we are working hard to make them grow.
Feel free to share your experience with integrating UX research within your organisation.