Nutrition App for Clarity, Habit-Building, and Smarter Tracking
My role was to redesign and modernize the existing nutrition application by refreshing the visual experience, improving friction-heavy user flows, and introducing new product features. I worked across the full product experience — from analytics dashboards and recipe discovery to food group balance tracking and an intelligent food logging system. The goal was to transform a functionally complex product into an experience that felt faster, clearer, and effortless to use on a daily basis.
Platform
iOS and Android app
Duration
6 months
My Role
UX/UI Designer
Team
Founder/Full-stack Engineer, Project Manager,
key outcome
–35% reduction in time to log a complete, precise meal
key outcome
Improved food database structure
key outcome
Improvement in long-term food tracking consistency
about
I Eat Better (Food Diary) is a mobile nutrition platform that helps users understand and improve what they eat — not just count calories. The app combines daily food logging with deeper tools: nutritional analytics, recipe collections, food group balance visualization, and personalized meal recommendations. The redesign brought all of these into a cohesive, modern experience.
The platform serves both users who require highly accurate nutritional tracking for medical or metabolic reasons and everyday health-conscious users focused on building better habits, managing weight, or increasing awareness around their nutrition. While their motivations differ, both groups share the same challenge: maintaining consistent food tracking without the process becoming mentally exhausting or overly time-consuming.
Case Study
Redesigning the Food Logging Flow: From Manual Entry to Smart Predictions
Food logging is the most repeated action in the app — users do it three or more times a day, every day. It's also where the product was losing people. The existing flow was search-based: find a food, add it, repeat for every ingredient, addition, and variation. For a simple meal, manageable. For a full meal with cooking methods and additions — the kind of precision that medical users needed — it was exhausting enough that users were either abandoning logs mid-entry or completing them inaccurately. The redesign of this flow was the highest-leverage UX problem in the product.
challenge
The core challenge was that precision and speed typically work against each other — accurate tracking usually requires more manual input, more decisions, and more time. For users managing medical conditions or strict nutritional goals, even small details like oils, sugar, or cooking methods mattered, yet entering them created friction that often led to skipped steps or abandoned logs.
The goal was to design an experience where precise logging felt just as fast and effortless as basic tracking by shifting more of the cognitive work from the user to the interface itself.
Key insight
The search-entry model required a separate action for every component of every meal. For any meal with more than two or three items, the cumulative cognitive load caused users to abandon the log entirely — or to simplify it into something inaccurate.
Manual entry was the #1 drop-off point
solution
We worked with the data team to build a custom suggestion engine that operated on two levels: pattern-based (what most people add to this food) and personal (what this user specifically adds, based on their history). The UX goal was to surface the right suggestion at the right moment — before the user had to think about it.
before
1
Search "coffee"
2
Select match
3
Set quantity
4
Search "sugar"
5
Select match
6
Set quantity
7
Search "cream"
8
Select match
9
Set quantity
after
1
Search "coffee"
2
Select match
3
Set quantity
4
Select suggested addition
5
Set quantity for addition
Over time, the system learns individual patterns. If a user always adds honey to oatmeal, honey moves to the top. If they consistently boil their eggs, "boiled" becomes the default. Returning users experience a flow that increasingly mirrors their actual behavior — reducing even the suggestion-scanning step.
