05
Year 3, Jul to Sep 2026
Mobile app, UX and UI
Mettle
Workouts, diet and activity, in one app.
When
Year 3, Jul to Sep 2026
What
Mobile app, UX and UI
Tools
Figma
My part
The goal, research direction, design decisions and audit.
There isn't an app on the store that puts every workout, a diet planned around your specifics and activity logging in one place, so people end up running three apps that never share what they know. Mettle is that one app. It's designed in dark and light themes, with a wired prototype of about 150 interactive nodes.
Why I made it
There isn't an app on the store that includes all workouts, a diet with your specifics, and activity logging together. That gap is Mettle's reason to exist.
Before designing, I checked the gap. I compared eight of the most recommended apps, sorted published findings on why people quit fitness apps into an affinity map, and walked a typical week through a journey map.
The research points at effort: across health apps, a median 70% of users stop within 100 days. So the goal became one place for everything, with as little typing as possible.
Who already does what: eight popular apps, each strong at one job.
Affinity map of published findings on why people quit fitness apps.
A week with three apps, and with one.
The screens
Scroll, drag or use the arrows to move between them.
The point was to ask whether making something easier can make people care less.
If you make a task effortless, people are more likely to keep doing it. But strip away too much and the app starts to feel like it isn't paying attention to you, which can push people away for a completely different reason.
How it moved
Competitors
Each popular app is strong at one job: food logging, strength logging or activity. None covers all three, so data ends up split across apps.
Affinity map
Findings fell into five groups: too many apps, logging is work, plans don't fit me, losing the thread, and cost and trust.
Journey
A week of decide, set up, train, eat, move and check progress, with the pain at each step and where one app can remove it.
Screens
Train, Fuel, Log and Profile map to that journey: today's session with sets and rest built in, meals planned for veg or non-veg and logged in a tap, activity beside both, and one weight trend against everything you logged.
What's missing
This is secondary research. I haven't run interviews or tested with real users yet, and that's the next step.
The work
Learning
Heuristics only make sense once you use them. I found dead ends only by clicking through my own prototype. I also noticed I keep postponing structural work, like reusable components, in favour of finishing screens.
Design
Audit, document, fix, re-audit. I logged contrast failures, wrong macro maths and small tap targets, then corrected them. My definition now balances ease with control: making something effortless isn’t always better if the user stops feeling in charge.
Experience
A lifter opens the app at the gym, sees today’s session, logs it in a tap and checks fuel. It has only been tested by me. The central question is still open: does removing daily friction raise adherence, or remove enough control to backfire?
What others showed me
I leaned on research here: protein and training studies like Morton et al. 2018, and Nielsen's heuristics to audit the app. No real users have tried it yet, so I've only tested it against rules and against myself.
Strongest at
Systems and rigour.
Still growing
Reusable components and testing with real users.
Next
A proper component library, then test the logging action with real users over several weeks.