- Role
- Solo: research, UX/UI, build
- Platform
- iOS (iPhone)
- Status
- Beta on TestFlight
- Scope
- Research · Strategy · UX/UI · Design system · AI-assisted build
At a glance
Symio in four numbers.
1 tap
to log a bad day. Saved instantly, no confirmation screen.
7
condition profiles with symptoms already set up, including “your own”.
6
sign-up and account screens removed from the first design.
0
accounts, servers or trackers. Health data never leaves the phone.
The problem
It started with a friend who kept forgetting.
Sarah lives with a chronic illness. During one treatment her doctor kept asking precise questions:
- How many times a day?
- Since when?
- Was it worse before or after the new dose?
She couldn't answer. She had felt every symptom, but by appointment day the details were gone.
That isn't a memory problem. Chronic illness is tiring, and symptoms blur together. During a flare, writing anything down is the last thing anyone will do. So the person who knows the illness best often gives the doctor the least reliable account of it.
Research
What I learned.
Discovery came first: I talked with people living with chronic conditions and benchmarked the existing tracking apps. The aim was to understand how people actually live with their symptoms, not how an app expects them to.
- 1
People track in bursts, not every day.
Tracking is intense when a new treatment starts, sometimes daily for months, and during flares. The rest of the time it barely happens. A tool that punishes gaps loses people.
- 2
Typing is a barrier when you feel unwell.
People wanted to tap buttons, not write sentences. Every keyboard interaction is a reason to skip a day.
- 3
The real question is “Is this treatment doing anything?”
Users wanted to see patterns, like pain peaking before their period, and to compare before and after a change.
- 4
Existing apps make you choose between powerful and pleasant.
The complete ones are heavy to set up. The pleasant ones aren't built for medical use.
- 5
It has to feel kind, not clinical.
“Not another calendar.” Encouragement mattered, and guilt drove people away.
Benchmark
Powerful or pleasant. Symio aims for both.
Bearable and CareClinic are comprehensive but slow to set up. Daylio is simple and enjoyable but not made for medical use. Medisafe is about medication reminders, with basic symptom tracking.
- Daylio: Simple and enjoyable, but not made for medical use.
- Symio: Aims to be both pleasant to use and medically useful.
- Medisafe: About medication reminders, with basic symptom tracking.
- Bearable: Comprehensive, but slow to set up.
- CareClinic: Comprehensive, but slow to set up.
From insights to requirements
Three jobs, four principles.
- Job 1
Track during a crisis without effort
One-tap logging, symptoms already set up for your condition, no keyboard.
- Job 2
Never miss, and always prepare for, an appointment
Reminders, a preparation checklist and questions to ask.
- Job 3
Bring real data to the doctor
Trends, before and after treatment comparisons and a one-page summary to share.
Fast over complete
Every input screen can be finished with a thumb.
Kind, never guilt-inducing
A day without a log is just a day.
Show, don't interpret
Symio shows numbers and curves. The doctor interprets them.
Private by design
No account. Data stays encrypted on the phone.
The solution
Six features, each built for a bad day first.
The app is in French, written in the informal “tu” to sound like a friend, not a clinic.
1 · Check-in
A check-in built for bad days
Five faces, one tap, saved instantly. On better days, a full check-in adds each symptom on a 1–10 slider, a body map to locate pain, sleep times and an optional note.
- Low-mood faces are soft pink, never alarming red.
- Every body zone has a touch target of at least 44 pt and is announced by VoiceOver.
- Every step has a way out: “Pas aujourd'hui”, “Pas de douleur, suivant”.
2 · Onboarding
Personalised in under a minute
Pick a condition (IBD, endometriosis, fibromyalgia, MS, lupus, long COVID, or your own) and Symio pre-selects the relevant symptoms. You can add or remove some, but you never have to build anything from scratch.
3 · Treatments
“Did my treatment change anything?”
For each treatment, Symio compares every symptom before and after the change, on a 0–10 scale. It shows the numbers and a dotted line on the chart where the change happened.
No green or red, no “it works”. A difference of 0.7 points is stated as a difference, and the patient and the doctor decide what it means.
4 · Reports
Reports that speak the doctor's language
Weekly, monthly or custom trends for every symptom, with lowest and highest values and treatment start and stop markers. Missing days show as “not logged”, never as 0.
One button turns the period into a one-page PDF, generated on the phone and shared through AirDrop, email or print.
5 · Appointments
Appointments, prepared
A care team, upcoming visits, a reminder the day before and the same day, a preparation checklist and a list of questions to ask. The home screen shows the next visit and how much of the preparation is done.
6 · Privacy
Privacy people can feel
No sign-up, no email, no cloud. Health data lives in an encrypted database on the iPhone, works fully offline and can be locked with Face ID. There are no analytics or advertising SDKs, and notifications never show health information.
Deleting everything is one clear screen: what will be erased, an offer to export first, and a typed confirmation.
Design decisions
What I removed, and why.
The first complete design (v6) looked good and tested the idea. Building it for real meant checking every sentence against the principles. These are the cuts that shaped the product most.
Six sign-up and account screens
“Your data stays on your phone” became the core of the onboarding. With no account to create, six screens disappeared.
The “pattern noticed” cards
They looked helpful, but they implied a medical judgement. Symio shows numbers and curves, and the doctor interprets them.
A separate “crisis” button
It split logging in two and added a choice on the worst days, exactly when nobody wants to choose.
The “peaks” section of the doctor's PDF
It labelled days as flare-ups for the doctor. The summary now shows the data and leaves the reading to the doctor.
Design system
A calm notebook, not a medical chart.
Cream backgrounds, soft white cards, rounded type and gentle lavender and pink. Text starts at 16 pt with large touch targets and one primary action per screen, so it stays readable when you feel awful.
Colour
- Cream#F7F5F2
- Primary#5B5BE0
- Pink#FBE3EE
- Ink#1E1B2E
Intensity scale of the history calendar, 1 to 10. Violet gets deeper with the day's highest rating, without ever turning red.
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
Type and accessibility
Comment tu te sens ?
Outfit for titles · Figtree for text
- Colour never carries meaning on its own.
- All text meets a 4.5:1 contrast ratio, in light and dark.
- Full dark mode, Dynamic Type and VoiceOver.
- Touch targets of at least 44 pt, including the small body zones.
How it was built
One designer, from insight to a working iOS app.
I designed and shipped Symio alone with an AI-assisted design-to-code workflow. I wrote the product and design rules, and every change was checked against them. That let me test real features with real people instead of static mockups.
- 01
Research
Interviews and benchmark
- 02
Product spec
Scope, principles, what V1 won't do
- 03
Design
Screens and design rules
- 04
Working rules
No network calls, no hard-coded text
- 05
Build with AI
One task at a time, each decision logged
- 06
TestFlight
Real iPhones, real users
Words matter for App Review too.
For the appointments feature, I chose the wording carefully. “Booking” would suggest Symio contacts medical offices and shares data with them. It doesn't: you write down an appointment you already made.
- Avoided
- Book an appointment
- Used
- Appointment planner and reminders
Validation survey
Checking the idea with patients.
Once the app worked, I surveyed 7 people living with a chronic illness: lupus, endometriosis, coeliac disease, Ehlers-Danlos syndrome, asthma, anxiety and depression, and pancreatic insufficiency. I wanted to test the discovery insights, and Symio itself.
October 2026 · 7 respondents aged 21 to 32 · All agreed to the use of their answers. These are signals, not statistics.
7 of 7
rely on memory to keep track of their symptoms.
4 of 7
only write symptoms down during a flare.
6 of 7
have never used a health tracking app.
4
health professionals per person (median).
5 of 7
take a treatment several times a day.
3 of 7
say tracking demoralises or stresses them.
What it confirmed
People track in bursts.
4 of 7 only log during a flare, and just one logs every day. A one-tap check-in and the option to fill in past days fit how people really track.
The real question is the treatment.
One participant's missing feature in other apps: “trends, a link with new medications or doses, and a nice interface.” That is Symio's pitch, almost word for word.
Care is spread across many people.
A median of 4 health professionals, and 4 of 7 prepare a list of questions. The care team and the questions to ask are features people already do on paper.
It has to be kind.
3 of 7 say tracking demoralises or stresses them. The gentle tone is a requirement, not a style choice.
What it challenged
The barrier is usefulness, not effort.
“Not useful” is the most cited reason for not tracking (3 of 7), and everyone would spend at least a minute a day. Next step: show a first useful view before 7 check-ins.
Streaks don't fit people who track during flares.
The home screen celebrates regular logging (“2 more days to your 7th day”). For someone who logs only in a flare, that turns into a reminder of what they don't do. Next step: make the milestones neutral or optional.
“An empty day is a normal day.”
One participant wants unfilled days to count as normal. I keep “not logged”, because a doctor needs to know when data is missing. Next step: a one-tap “Nothing new” that marks stable days, one or several at a time.
“The app wasn't practical and I couldn't see how it could help me.”
“Assume that if I don't fill in a day, it was normal.”
“Precise tracking wouldn't bring me much. It might even stress me more.”
Results
The beta is running.
Testers are using Symio on TestFlight now. I'll publish the results here when the survey closes. These are the goals I set before the beta:
- for a full check-in
- < 10 s
- of testers still logging after 30 days
- 25%
- testers bring the PDF to a real appointment
- Doctor
What's next
Next steps, ranked by the survey.
- 01
Medication reminders, including “as needed” doses
5 of 7 take a treatment several times a day
- 02
Symptoms grouped by body system for conditions outside the list
5 of 7 had a condition that isn't pre-configured yet
- 03
A first useful view before 7 check-ins, and neutral milestones
“Not useful” is the most cited reason for not tracking
- 04
A home screen widget for a check-in without opening the app
One participant's life-changing feature
- 05
A “difficult days” filter in the history
One participant wants to find every flare quickly
- 06
Android
3 of 7 use an Android phone
- 07
An emergency card, English and German versions, Apple Health, optional encrypted sync, and symptom lists reviewed with clinicians
From the original roadmap
No participant would pay for the app. Keeping Symio free, with no ads and no data sold, fits that, but how to fund it is still an open question.
What I learned
Three lessons.
- 1
The best feature is often the one you remove.
The crisis button and the “pattern noticed” cards both looked helpful. One added friction and the other implied a medical judgement.
- 2
Designing for bad days makes the app better on every day.
A check-in you can finish during a flare is also one you'll finish on a busy Tuesday.
- 3
Privacy is part of the UX, not a legal footnote.
“Your data stays on your phone” became the core of the onboarding, and it removed six screens.
Symio · Case study
Up next