Update: Just some days after the publication of this case study Trenitalia has updated its application. Happy to see a lot of improvements!
My Role: UX Designer | Duration: 5 Weeks | Year: 2020 |
Tools: Figma and Grammarly
Summary and Techniques
Through a design thinking process, I’ve gathered a lot of information about what users need and which are their pain points. So:
- A usability study was useful to gain insights that I didn’t expect;
- A heuristic analysis gave me the chance to systematize my findings;
- The competitive analysis has suggested me some clean solutions to organize contents on the page;
- In information architecture, more pieces have found their place;
- In the end, my process led to the solution showed in wireframes.
Two months. That’s the time that I needed to write the Trenitalia app’s case study. Part of my journey to come back to UX is something that may help me become a more effective professional and a better person. So let’s go.
Even if it’s super evident to my eyes how many problems are in the Trenitalia app, I know the context where these kinds of issues come up. First things first: this is not incompetence, at least for people who in Trenitalia make real work.
To the opposite, probably many people know very well what they have to do in their jobs. In big companies, politics and legacy are the worst enemies you’ve to face it.
In 2018, the app’s previous version had three stars in reviews in the App Store and Play Store. With the new version of the app, things are even going slightly worst.
I guess that a user-centered design would be beneficial for the application.
My first step was to empathize with users. That’s why I set up a usability study, involving family and friends, to better understand struggles and pain points from a user point of view.
Every session was audio and on-screen recorded to be sure that I didn’t lose any detail. Plus I collected information about errors during the tasks, annotating:
- Slips, unintended actions (like a typo) while goal is correct;
- Mistake, what is wrong is the goal. It’s an error even when the user reach the goal;
- User interface (that causes directly) errors.
For each task, I also recorded:
- Completion rate;
- Time needed.
And the satisfaction level, overall, for each participant.
Here an infographic to have a quick overview.
On task #3, the most complex task, Buy a ticket. I’ve recorded:
- One user failed it;
- High times to complete it;
- A lot of errors.
Two stories are useful to understand critical things that I learned and I didn’t know about users.
User #1 had chosen a trip for one person instead of two people without noticing. So, selecting seats, he was not aware of why he can’t pick a second one. Only through a process of trial and errors (5 in 3 minutes), he understood the reason.
A new inspiration for my design!
For me, a consequence of this insight was to:
- Show on each page of the step the chosen made;
- Give the chance to change or correct the user’s choice.
User #5 failed the task as he was buying a ticket for the day next. He was so focused on his aim to ignore any text alerting onto this.
I guess you’re curious how it was possible. Check out by yourself in this video.
As you may see, there’s a sentence under each card, which means: the solution refers to the next day. That’s not enough because there’s a reminder of what he’s buying on the next page. Unfortunately, a green tip almost covers the text.
The main problem here is a lack of prioritization between contents that fight each other for the user’s attention.
A new inspiration for my design!
In my solution, buying a ticket for a day different from your first choice is possible, but the user has to take a different path. On the listing page, he has to press a Next Day button.
To progress in my process and build upon the usability study’s insights, I moved on to heuristic analysis with a psychological approach. In this way, I had the chance to:
- Check every aspect that doesn’t work in the app;
- Put all the information in a full picture.
For each control, there’s a severity check to define if the control passes or fails. Zero means no problem, while 1 is a cosmetic issue to reach 4, a catastrophe.
For example, thinking of the user #5, a catastrophe is to buy a ticket for the wrong day. Here, two controls fail related to the heuristic People make mistakes:
- The system has to require user confirmation before taking actions with severe results in case of error.
- It’s better to break up error-prone tasks into smaller chunks.
My next deliverable was a competitive analysis to generate ideas: I decided to focus on the buying process. Analyzing competitors in Italy (Italo) and abroad (Trainline and Trainpal) highlighted best practices.
As a result, it was clear, which was the best way to organize things when there is a lot of information to show, like in the listing or the service level page.
Moving on, I’ve simplified navigation, mainly:
- Fields to start research are available immediately in the home;
- Trip details are now in each step of the buying process, always at hand, if needed;
- Information is more comfortable to reach. Like in the case of Stops, fermate, each content is in the first level of the information architecture.
Only after all these pieces of information and insights were collected, I was ready to draw wireframes and feature how Trenitalia app could be. In respect of some of Nielsen’s heuristics:
- Match between system and the real world. Language is simple and friendly.
- User control and freedom. The user has everything needed at hand. He may choose the best solution for his needs, like add more passengers later in the process if desired;
- Consistency and standards. Primary and secondary buttons, background, header, footer are always the same and used consistently throughout the app.
- Error prevention. I’ve introduced a Next day button on the results page. In this way, the user knows that what will see regards something different from what he has chosen on the search page.
Just a quick preview of my work, but if you want to check more, take a look at the prototype!
- Not only five users + one designer are far more valuable than one designer that works alone. I’d say that they cost less too: better quality means less rework and in the end a saving.
Covid 19, something that almost nobody expected, has arrived and human beings are watching in the face of their worst fears. My wife works as a physician in palliative care: as a family, we feel very fortunate.
So, when things improve, I hope to run a second usability study to measure improvements between the previous version of the app and my proposal.
There are two persons that I’m very grateful for their support and help:
Do you want to discuss more details on this case study?