Providing a better user experience for Rio's public transport kiosks
The context
This client is a private company that provides electronic smartcards used to pay for public transport in the Rio de Janeiro state.
The smartcards can be used to take the bus, metro, train, etc. and are sold at the brand's stores or from kiosks at selected stations. Users can also top them up on the mobile app, the website or the kiosks.
They hired a design agency because they wanted to improve the system and ergonomics of their kiosks.
The challenge
This is where I came in, to conduct research for the client and identify opportunities for improvements in the interaction of the users with the kiosks.
So some of the general goals we had were to:
Understand how people use the kiosks, their gestures and attitude, mood, etc
What do they do? Do they top up their cards with cash or card? How do they go about it?
What difficulties do they have? and how do they try to solve them?
The process
Desk research
With that in mind, I obviously had to start with some desk research for a deep dive into the context and better understand what we already knew, our hypotheses and questions.
Through a CSD (Certainties, Assumptions and Questions, in Portuguese) matrix, we listed what we were certain about, our assumptions and the questions we needed to answer through research.
CSD (Certainties, Assumptions and Questions, in Portuguese) matrix
Reviews and complaints on Reclame Aqui, an important Brazilian website for complaints against companies about customer service, purchase, sale, products and services.
General information about the client, the brand, its mission and goals, etc.
Benchmarking
I analysed direct (and indirect) competitors to study and verify market trends and raise ideas for possible ways to innovate and add value to the project. The players mentioned by stakeholders were analysed to identify new trends, references, best practices and opportunities.
Players from other sectors and markets that could be used as inspiration for the project were also analysed. Furthermore, I did a competitive analysis to gain insights into what they offer to the clients.
The preliminary analyses of heuristics and usability helped me get a picture of some of the main issues with the current system and come up with some hypotheses.
Many screens didn't pass on contrast tests, presenting texts that could be unreadable for many people.
The system generally lacked consistency in elements and assets
Stakeholder interviews
The team had a few rounds of interviews with stakeholders to better align expectations, understand different perspectives of the problem and raise limitations or concerns we might face.
We spoke with 4 people from management positions and 3 people who are on the field daily at points of great circulation providing support to users of Rio Card ATMs.
The main points raised were:
The ATM system is slow
There are many malfunctioning machines
Barriers (tech) to app adoption
The ATMs and their system are not accessible to those who do not read or read little
Customers do not try to solve problems independently
Elements that overlap the main screen confuse the user
Interface with too many elements overload the user
Error messages are not clear
The ATM system involves a considerable learning curve
Difficulty understanding what needs to be done in the system
The impressions raised by internal stakeholders may not always correspond with those of customers/users.
That is why efforts to bring the customer/user's vision into the company are so important to provide a good experience and, consequently, generate more business.
At the same time, it is interesting to note that great suggestions can emerge from these sessions, as well as points that coincide with the rest of the research.
Field research
Ethnography
By observing the users in the field, we seeked to analyze their behaviour in a natural context without interference. Aspects such as posture, body and facial language were observed, if they seemed to have any difficulties or doubts, how they interacted with the ATMs, etc.
The stations observed were: Carioca, Central do Brasil, Jardim Oceânico and Largo do Machado.
The main points raised at this stage were the following:
The ATM system is quite slow
Malfunctioning machines disrupt the flow
Current system has too much information on the screen
Difficulties when using the ATM
Bad touchscreen or the person missed the button
Users hesitant where to click
Confusion with machine inputs and outputs
Users need to keep reading the options and instructions at every step
User behavior
Most users made their journey alone
Those who were accompanied used the machine together, helping each other
Many already look directly for a promoter present to ask for help (especially at Central do Brasil). They do not seek another way before
Experienced users use ATMs confidently
People don't pay attention to signs and signs
Patterns in every season
User interviews
With user interviews, we wanted to better understand the pain points and needs of users, as well as their contexts, stories and relationships with the challenge and theme of the project.
A total of 28 individual interviews took place.
Respondents were approached in the stations, close to the client's ATMs located in subway stations, BRT and brand stores in Rio de Janeiro.
Here's what stood out from talking to users:
Slowness of ATMs hinders users
Machines broken or with system error make the journey more difficult
Poor system efficiency and usability
Confusion with the physical part of the machine (slots for the card, money, receipt, etc)
Most people use the ATMs because it is their only option
Behavior varies greatly, according to age, purchasing power, region where users live/work, how they prefer to top up the card
Customers feel there should be ATMs in more locations
Some wish they could pay on credit (at the time of this study, this was only available at station Largo do Machado)
There is a learning curve with the system
Usability tests
We also did some usability tests to analyse the behavior and interaction of the user with the proposed solution and its efficiency.
The test consisted of tasks that had to be completed by users, randomly distributed between: card purchase or recharge, with payment in either cash, debit or credit.
The tests were applied to 12 people with different profiles.
During the execution of the tests, some metrics were collected:
Success rate (whether users were able to complete the task)
Error rate
Subjective user satisfaction
Note: considering the test conditions and context, it was not possible to accurately measure the time required to complete the task.
We did 2 rounds of usability tests, iterating between them. We were able to improve the overall satisfaction from 60% to 72% and the task success rate from 40% to 72% of people who successfully completed the task with no errors.
The designer started the work with a low fidelity prototype
V1 for the usability test. From left to right: home screen, instruction screen for reading card, instruction screen for cash payment.
V2 for the usability test. From left to right: home screen, instruction screen for reading card, instruction screen for cash payment.
Ergonomics
Suggestions for the physical ATM machines, based on field research findings:
We organized the signage into 3 categories, namely: (1) indications of action location on the machine [purple], (2) Logo [blue], (3) Support signs [pink].
For each category, it would be interesting to follow a color pattern, with the following suggestions:
(1) For indications of action location [purple], it's recommended that the adhesive follows the same color as the animation on the screen, so that the user can associate the action between the system and the machine.
(2) Today, the labeling has the background in blue and the logo in white, we can keep it that way.
(3)Support signs [pink] can be chosen by the client's marketing team or requested separately for the agency.
Additionally, we also recommended smaller and specific signage to be either removed, kept as is, change location, or kept with some changes.
Conclusion
The team provided all deliverables to the client, who will now test the dark mode vs. light mode prototypes live on the field to determine which works best and later implement on all ATMs.
The materials were well received, with some minor requested modifications to best adhere to specifications the client needs to comply with. I have since switched companies and unfortunately won't be able to learn of the final results.
.jpg%3Ftable%3Dblock%26id%3Db7952fb6-fce7-4ef6-bd3a-bfae88a396c0%26cache%3Dv2?width=1500&optimizer=image)
