Project
A company subsidiary focused on B2C sales of (craft) beer and beer draft systems.
Duration
Sept 2020—February 2021
My responsibilities/role
UX Research Lead
Technologies
Zoom (In-Depth Interviews), Otter.io (Transcripts), Miro (Analysis, Workshops), Jira (Project Management)
Context
Beerwulf lacked qualitative data about its customers. I proposed to conduct a qualitative study of the Beerwulf customers. Subsequently, I was granted the responsibility to lead research for the UK market.
The Internship
During the first lockdown of the COVID-19 outbreak, I was coming through a difficult phase of my life, and I felt I need a change in both my personal and work life. As I was still a working student, I thought taking advantage of internship opportunities could be quite interesting to explore new environments abroad as well as take a sneak peek of working in global, game-changing services. After the never-ending preparation of my portfolio, writing to countless companies and a couple of interviews, I finally received an acceptance e-mail from one of the greatest companies a Czech citizen can dream of — Beerwulf, Heineken Netherlands.
Coming to Heineken
Right after entering the headquarters building, I knew this was the best choice I could’ve made. The UX team made sure not only to settle a warm and welcoming 1-on-1 with every of my team members as well as to be introduced to the rest of the other teams. At that time, I didn’t know that in the end, it was going to be pretty much everybody in the company including most C-level executives.
Feedback on existing stuff
My introduction to the service and products Beerwulf offers happened in a form of a first-person usability test of our website and internal processes. After an in-depth analysis of found issues and proposition of possible solutions, I was finally ready to start working on my own project.
Getting responsibility for researching the British market
One of the issues identified during my analysis was a lack of qualitative data about our customers. Moreover, the team had the best database of quantitative data I’ve ever seen from millions of different sources but had no idea about who these customers, as human beings, actually are. They didn’t know if their typical customer is a 50yo financial specialist who likes to drink a cold one while resting after a tough day or a 25yo student who needs a beer for their Friday party. They didn’t know their age (except being 18+ ofc), didn’t know their interests, life struggles, goals, pain points … only information from the CS team based on the negative feedback they’ve encountered. At this point, this was quite a lesson. The whole time I studied service design, our teachers were emphasizing the importance of complementing both qualitative and quantitative research during the design process. This showed me that no organization is perfect and not everything learnt in school is always so obvious in a monumental living entity such as a giant corporation.
Once I presented my opinions to the manager, he responded in a way I did not expect. “Why don’t you take over the responsibility of starting the project of qualitative data collection as your internship project? Let’s start with the British market.”
Aaand so my internship project began…
The Team
Designing the Research
From the very start, I was aware that it was going to be an enormous project that is supposed to last way longer than only during my internship and that the base that I set is going to affect quite a lot of decisions throughout the whole company in future. With this in my mind, I knew that the design of the research has to be prepared thoroughly and comprehensively.
I have started by creating the draft research design that I later discussed with a couple of senior researchers to expose potential flaws that could affect the research in future stages. After polishing and tweaking a few methods and techniques, I finally come up with the final proposal.
The biggest issues that we discussed were regarding the ratio between qualitative and quantitative metrics used. As I’ve mentioned earlier, Beerwulf used to be a heavily data-driven company, thus they already had a surprisingly large amount of quantitative data collected from their data team. The goal of my research, therefore, was to explore the qualitative side of our audience, merge this data with the quantitative part and come up with an overview of our customers in the UK market.
Once all the tweaks were done, I also presented it to the manager, who gave me his own feedback to match the company’s processes and I received the approval to start with the research.
Perks of Researching
As we were not sure about the differences between our customers and the methodology we designed at the beginning, we decided to approach the research project itself in a design thinking way. We divided the whole research into more granular parts to be able to learn from imperfections in past stages and improve these processes and build on them in future.
As the company had basically no qualitative data, we weren’t sure how to target our users properly, to aim at all important types of the whole area. Also, we needed to divide the research into more granular steps so that we can grasp all the different parts of the research more easily.
The only thing that helped us in the beginning (and thanks God for that) was that Beerwulf sells two main types of products — bottled beer and draft beer kegs and machines. As we knew from the data we already had (obtained from CS, data, and user testing team), the behaviour of these two types of users differs practically in every aspect of their buying journey. From the products, they are looking for, to the brands they’re interested in, as well as in the preferences when choosing the delivery options. We thus had a hypothesis that we will need to talk to a sufficient number of customers from both groups to cover all user types. Because of that, when contacting our customers, we were choosing the same number of users from both groups.
Analysis of such a huge market
Once everything was planned, we started conducting semi-structured in-depth interviews with our customers. By the way, this was the first time I used the Calendly tool, and I can’t live without it anymore :)
Even though I’ve had quite vast experience with working in international teams from uni, this was the first time I conducted the whole research process in English only. Fortunately, my initial fears of understanding the true Scottish accent were over just after a few first interviews as I found out that I’m capable of more than I expected and the only thing you sometimes need is to calm people down, and eventually ask them to stop running with their dog when speaking to you :D Of course, all these funny situations only added some additional joy to the whole process, and I not only enjoyed the process itself but also knowing more about the whole beer culture in the UK, how much can a good pint of beer be for some customers and what they value about our services the most. These were really the parts that sparked some joy inside me.
Once I finished talking to all of the respondents, I started with the analysis of the obtained information. Firstly, I finished the transcription of all the interviews (using Otter.io + manual correction of typos).
After this most painful part, I could finally move to the part that I enjoy the most which is the analysis itself and unveiling unknown patterns across our customers.
As we didn't know much qualitative data about the customers, it was hard to estimate what coding labels should I use for the coding of transcribed interviews. Therefore, I decided to use the open coding method.
After that, I started copying all the important quotes to begin with affinity mapping all these findings. All the quotes were written on post-its on which every respondent had their own colour of post-it allowing me to backtrack the context of the quote if needed. Also, this provided anonymity to the respondents when sharing findings with other teammates, as I was the only person having access to matching pairs between colours and transcripts including names.
For affinity mapping of these findings, I decided to choose a bit different approach to those I was used to in previous projects. As I knew that some of the collected data is nominative and some ordinal, and I needed them to be compatible I chose to create an affinity mapping workspace having two parts - One had ordinal characteristics and the other nominal. The ordinal part consisted of axes of two poles (extremes for every characteristic) and post-its with quotes were put on these axes to later form piles with similar behaviour or thoughts. Nominal had characteristics which have been grouped in free space just based on the similarity of these quotes.
Affinity Mapping Sketching
The first iteration of Affinity Mapping
This method allowed me to see an incredible amount of data easily visualized. One problem was that it was hard to find patterns of our typical users to later form personas. Summing up the biggest piles and forming personas just based on the most often found responses would sound tempting and easy but that would produce imaginary personas whose characteristics don’t make sense together. These characteristics have been very often mutually motivated and I needed to take into consideration who said it and what other characteristics he or she had.
To be able to untangle this complex situation I decided to create a contingency table of every respondent and count how many times has this respondent been on the same characteristic pile with every other respondent, in other words, how many times has he or she had a similar response as any other respondent. As we knew we wanted to have only 5 first personas for now that we’re going to study deeper and potentially create a more granular classification in future, it was then very easy to find them using this table. I basically had to look for the cell that had the highest number (highest level of similarity), compare how many respondents are connected this way together and based on their mutual characteristics form personas.
Contingency Table
Once this part was done, the only thing left was to present the outcomes to the stakeholders. Because my manager was super content with the process and results of the research he wanted me to also experience presentation to the C-level management. Few moments before the final presentation he told me that the CEO (Cindy Tervoort) and CCO (Krijn Jansen) of Beerwulf, Heineken are going to be present too. I'm not gonna lie, my legs started to shake.
In the end, the presentation was way better than I expected, all of the stakeholders complimented my work, and I finally knew that all the stuff I’ve worked this hard on, had a real purpose and will actually influence the future directions of this incredible company. Moreover, they invited me to present the findings to most of the teams of the company to instil the user-centred approach to the company's DNA .
Moral Dilemma
At the same time, at this point, I started to think about the moral aspect of the whole internship I am participating in. On one side, I really enjoyed working for a super open-minded company, where teammates are just so much fun to work with and whose products I love during the evenings with my friends. On the other side, I kind of blamed myself for working in the alcohol industry and supporting drinking which I personally feel like it is a way bigger problem in today’s world than we admit. It worried me more and more as I realised that we tried to enhance peoples’ drinking habits.
At the end of the internship, the whole company participated in a design workshop forming the company’s vision and future. I told myself that this is probably the right moment to speak out about my thoughts. I unfolded my concerns and surprisingly, the top management took my points very seriously and chose sharing drinking consciousness and its problems awareness as one of the core pillars of the company’s vision. This was an eye-opener for me. I knew that Beerwulf is an open-minded company and cares about its employees' beliefs. However, this act of self-awareness and responsibility for peoples’ lives has proved to me that this is the approach I expect from my future full-time employer.
Outcomes of the internship
Overall, working for Beerwulf has been an incredible ride. Not only that they were brave enough to give me an opportunity to lead an enormous project that now shapes the whole company’s future but also, they showed me how can a corporate-size company still have a human approach to its employees and still maintain fast agile development with strict deadlines and full calendar.
It's been an honour to finally learn Scrum in a larger company, to learn how to communicate between multiple teams as well as how to give a 40-minute talk to the CEO.
Even though some moments weren’t easy, in these 5 months I’ve learned more than in 5 years of college.
Thank you Beerwulf. You gave me a lot.
Teamwork