Gaia
6 weeks
Figma, Forms, Notion
UI/UX designer
I've lost count of how many times I downloaded a diet app, opened it, and immediately felt too tired to deal with it. The digital wellness market is saturated, yet many nutrition apps fail to retain users long-term. When I started this project I wanted to understand why so many people, including myself, gave up on it so quickly, and design a solution.
6 weeks
Figma
Forms
Notion
UI/UX designer
How can we improve the day-to-day experience of diet app users and empower them to maintain long-term, rewarding wellness habits?
Research
I ran a competitor analysis of leading nutrition apps to see how they approached logging, personalization, and motivation. To get a deeper understanding why diet apps lose users so quickly, I surveyed both loyal and lapsed dieters and got a clear result.
Research
Through the research I wanted to see how complicated logging actually is, so I asked some people from the research to screen record themselves logging one banana, and one table spoon of peanut butter from their go-to app. Here are some thoughts and questions I asked myself when analyzing the footage.
Confidence looked shaky. People didn't seem very confident in what they were logging. I noticed small pauses and backtracking.
Is a banana ever just a banana? Users got stuck deciding between grams, pieces, or size, and still got different results.
Everyone's seeing the same menus. It almost feels like the same app. Is there a way to make it more easy or delightful for users?
Is this really about data or about ease? If users get conflicting numbers anyway, maybe what they really need isn't "perfect", but quick.
Noom - 10 clicks, 50 seconds, 205 calories
Confidence looked shaky. People didn't seem very confident in what they were logging. I noticed small pauses and backtracking.
Is a banana ever just a banana? Users got stuck deciding between grams, pieces, or size, and still got different results.
Everyone's seeing the same menus. It almost feels like the same app. Is there a way to make it more easy or delightful for users?
Is this really about data or about ease? If users get conflicting numbers anyway, maybe what they really need isn't "perfect", but quick.
Noom - 10 clicks, 50 seconds, 205 calories
Findings
Ultimately, three critical insights rose to the top and set the direction for every design decision that followed.
It's exhausting.
Calorie tracking is exhausting. Users search a food database, go through identical entries, pick one, then start guessing with portion units like grams, ounces, cups. They quit.
It's unreliable.
Every new user lands on a default 1,200 calorie target. Users feel it instantly. This approach to coaching and nutrition made users question the app's credibility.
It's cold.
Users said they'd try an app that lets AI handle the hard tasks and serve guidance in real, human language. Users want replies that sound like a friend, not a food database.
It's exhausting
Calorie tracking is exhausting. Users search a food database, go through identical entries, pick one, then start guessing with portion units like grams, ounces, cups. They quit.
It's unreliable
Every new user lands on a default 1,200 calorie target. Users feel it instantly. This approach to coaching and nutrition made users question the app's credibility.
It's cold
Users said they'd try an app that lets AI handle the hard tasks and serve guidance in real, human language. Users want replies that sound like a friend, not a food database.
Ideation
Based on the research, I focused on simplifying core tasks like meal logging and accessing personalized advice. I mapped out the primary user journeys and user flows to visualize the app's structure and navigation.
A friendly, low-friction diet app companion that fits seamlessly into everyday life.
On the home screen, I intentionally kept core features familiar and intuitive, matching users' expectations from similar apps. One key decision was placing access to the chat right in the home screen, so users could jump into conversations quickly.
Result
By the end of the process, the experience felt noticeably easier, quicker, and more tailored to each user, just as I had hoped to achieve.
Noom - 10 clicks, 50 seconds
Gaia - 3 clicks, 12 seconds
Metrics
Here is the final impact at a glance.
reduction in taps needed to log a meal
faster meal logging experience
drop in user effort and mental strain*
reduction in taps needed to log a meal
faster meal logging experience
drop in user effort and mental strain*
reduction in taps needed to log a meal
faster meal logging experience
drop in user effort and mental strain*
reduction in
taps needed
to log a meal
faster meal
logging
experience
drop in user
effort and
mental strain*
*Based on NASA-TLX score & feedback from a small group of prototype testers.
Reflection
I've made it a habit to close each project with reflection on what worked, what didn't, and how I've evolved. I carry those lessons with me into whatever comes next. Here are my key takeaways from this project.
Seamless but human.
Gaia was built to make logging easier with chat based interactions. That means the interface must remain inclusive, culturally sensitive, and forgiving of typos or slang. The best tech meets people where they are.
Explore but learn when to stop.
Brainstorming dozens of wild ideas like cross-app integrations, stickers, digital pantry & gamification excited me and opened a lot of possibilities. But the real progress came when I zoomed in on the biggest pain point, meal logs.
The small things aren't small.
Sometimes, it's all about one less tap, or removing one unnecessary choice. Small choices compound and its often what elevates something from simply functional to genuinely great, trusted and loved by users.
