Rethinking Travel • Personal Project
Solving travelers’ pain points with an AI-driven mobile app
Role: UX/UI Designer | Timeline: 4 weeks | Tools: Figma, Figjam, Otter.ai
Background
For the UX/UI bootcamp I took through Columbia University, we had to design “an app that addresses a problem travelers face.”
To better understand travelers’ pain points, I interviewed six frequent travelers between 20 and 35 years old and learned…
Most travelers struggle with feeling unprepared when they arrive at a destination because they didn’t have the time or resources to plan ahead. They also mostly traveled with others and often found themselves frustrated by group indecision.
Solution
Design an app that aggregates destination information from vetted sources to help travelers efficiently prepare for their trips.
TL;DR
Don’t have time to read this whole thing? Here are the main takeaways.
-
(1) In designing an app to solve all of a traveler’s problems, there was a lot of information I wanted to include, and I struggled to find the best way to present it without overwhelming users or accidentally hiding features. After countless tests, I landed on a card design that felt spacious and familiar to testers.
(2) This was the first app I designed, and I loved the challenge of branding it. The design was initially an homage to 1950s lifestyle magazines—since so many recommendations on the app would come from editorial sources—but I realized that the app’s bread-and-butter was truly its AI capabilities, and I landed on an appropriately modern brand identity.
-
I played around with microinteractions throughout the app. One lesson learned: time is of the essence. If these interactions take too long, the millisecond of joy they could bring users is lost to annoyance.
This was also a lesson in the Law of Proximity, as I tried to put related information together in a way that visually tells readers they’re connected. Spacing and design continuity are key here.
-
Moving forward, I’d like to continue refining the app’s layout and UI. I’d need to test the app in a real-world environment (would people be using it on-the-go or before they leave, and is the design conducive to that?).
If the app were to be brought to market, there’d also need to be a metric for establishing what a trusted source is when it comes to making recommendations.
Empathizing
Through my user interviews, I learned about travelers' planning habits—like what sites they visit when they do have time to research their trip and what's gotten them stumped in the past (“Is the tap water safe?” “How much should this taxi from the airport cost?”).
Why not create an app that answers all their questions in one place?
“Figuring out logistics—like phone cards, transportation, weather—is such a headache.” — Interviewee
“I hate not knowing what’s actually popular with locals.” — Interviewee
How might we help people spend more time traveling and less time sifting through Google results—without compromising their experience?
Brainstorming
Meet Bonnie
Bonnie is a travel app that uses AI to tell travelers everything they need to know before (or during) their trip—from the best rainy day activities to how to avoid a planned strike. It also lets users on group trips vote on itinerary activities, saving them time in the moment.
I wanted the app to have a human-sounding name, to make the AI element more approachable, and landed on Bonnie, a mix between bon voyage and la bonne vie.
Challenges
One of the biggest challenges I faced with this design project was finding the best way to package all the information I wanted to include. My first design used a tile system, but testers felt like it was too busy and their eyes didn’t know where to look.
I then landed on a card design, inspired by social media and competitor apps like Pluto. I wanted this app to be quick to use, since time is such a big factor for travelers.
But opening and closing multiple squares is more of a pain than scrolling, I learned, so the final design involves a mix: Key information available at a glance, with cards available to those looking for a deeper dive.
Iterating
Testing
Users were tasked with planning a trip to Amsterdam from June 9 to 15—but they kept overlooking the “Add dates to trip” function. To address this problem, I made the datepicker appear only after a user enters a destination into the search bar, to help draw their eye to it.
There was also a confirmation page after the “Add to itinerary” action that kept tripping users up.
What was intended to be a way of preventing user error—if they accidentally added the trip, for example, or hadn’t finished inputting information—felt redundant to users who thought they had already added it. So I readjusted. I put more emphasis on different tasks: (1) asking the group if they want to add it to the itinerary and (2) taking advantage of Bonnie’s smart date suggestions (e.x. Visit this museum on Tuesday, when it looks like it will rain?).
Branding
The design continued to evolve as I moved into high-fidelity iterations and performed more user tests.
The app’s style was originally a bit retro, inspired by 1950s lifestyle magazines, and it evolved into something more modern and future-looking, appropriate for an AI-driven app.
The destination page ended up being a mix of cards and card-less information, as I tried to reduce visual distractions, and the menu bar—now with a chat feature—became more streamlined.
Conclusion
This project was a lesson in user preferences. People want tons of information, I learned, but they don’t want to be bombarded with it. While my first design was crammed, user tests showed me that there are other (better) ways to present information.
N.B. One of the challenges this app would face if brought into development is the identification of which "vetted sources" Bonnie would scan from. There would need to be some trust metric—or maybe users who, say, trust Reddit over NYT, could add sites to their lists for Bonnie to scan.
