Stefánia Péter

User Experience Engineer & cat lover

Research case study: Mol Bubi

Context and background

I have participated in a User Researcher course called MOME Insight in September, 2021, a practical training that focuses on research, with Hungary's experienced research professionals sharing their knowledge. The course builds on the basics of UX, and is complemented by knowledge of commonly used research methods (interviewing, diary study, observation, questionnaires, usability testing), their practical application, and directly applicable advice.

Our research was about Mol Bubi, the most popular bike-sharing system in Hungary.

My Role

The participants were assigned in groups of threes, each group mentored by an industry professional. Group members shared the workload amongst themselves, each member conducting one interview. We put a survey together, did three usability tests, did some desk research and a first click test.

Since we didn’t have too much time, we split up some more complex tasks amongst ourselves. I was responsible for analysing the data from our survey, which was filled out by 118 people. Also had some minor part in creating the personas.

Timeline

The course was 12 weeks long, it started on 21th September, 2021, and ended in December in the same year.

Research statement and goals

We wanted to find out if the bike-sharing system in general has a future in Budapest:

We also checked how other sharing system, such as cars and scooters perform. Is there any connection between using a car and bicycle or the public transport?

Research methodology

We used several methodologies, such as desk research, user interview, survey, usability test and one-click test. To visualise the data, we used affinity diagrams and empathy maps. The results of the survey can be found here (Hungarian only).

Recruitment criteria and process

For this study, we were looking for a mixture of users and non-users. We did this because we wanted to make sure we got a comprehensive perspective of the service, and see how first-time users respond to the flow and experience. This helped us to ensure we weren’t just looking at power users who were "used to" the service.

We wanted to target 2 user groups by the following criterias:

Analysis and synthesis process

We recorded every interview, and took notes. We have analysed these with the help of our mentor and with the instructors. We put the facts on an empathy map, and used an affinity diagram too.

Picture of the empathy map
Empathy map
Picture of the affinity diagram
Affinity diagram

Usability test process

To test the process fully, we created a prototype of the mobile application in Figma. Since we had a bad weather and also COVID-19 restrictions were in order, we didn’t have the option to meet in person so we had to choose this method.

The participants shared their screen in Zoom, and followed our detailed instructions. We made up a little story to make them feel more connected to the test. We split up to the testing to three tasks:

  1. Sign up, fill out the required fields
  2. Explore the pricing options and buy a monthly ticket
  3. Find the nearest bike and rent it

One-click test results

The usability test suggested that the sign up was not obvious, but I didn’t fully agree, because one of our three participants had a technical difficulty, and after solving it, he was way too nervous to fully pay attention, he just rushed over the test. Two out of one was stuck a little bit on the process, so only one person doesn’t indicate that there are any issues finding the sign up button. After sharing the test, 75 people responded and 78% of them found the right button (the app has several buttons for reaching the registration form).

Conclusion

Most people like the bike-sharing system, because it’s cheap, easy to use and always available. They usually use it for going to, or coming home from a party or an event. Interviews showed that time was important for the users, so probably in these scenarios it doesn’t matter if the ride takes 10 minutes or 20, while when they are going to work, they want to get there quickly and in time.

The satisfaction with the service highly depends on which districts people live or work at, because in the downtown area, there are a lot more docking stations than in the outskirts.

People who use it regularly said that it’s really comfortable that you don’t have to take care of a self owned bike. No maintenance fees, don’t have to stress about how and where to store it.

Those who use it rarely mentioned several issues why they don’t really prefer the bike-sharing system:

Other transportation types, like shared cars, received positive feedbacks, but people don't combine it with other methods. The most negative review came for the scooters. People find them extremely dangerous and expensive.

Reflections

What went well:

Challenges: