Yagmur Şimşek Sönmez — I design things

Selected work ↓

Designer. Researcher. Crafter. PhD in architectural history, Master’s in visual communication design. Strong interest in visual identity. Wanna know me better?

Hi

Yagmur in a courtyard in Istanbul

Hi, I’m Yagmur.

I’m a product designer in Berlin. I design apps from the first user interview to the App Store.

  • Based inBerlin
  • SpeaksEnglish, German, Turkish
  • PhDHistory and theory of architecture
  • Master’sVisual communication design
  • Shipped4 apps on the App Store
  • Outside workI dance Lindy Hop

How I got here

I didn’t start in tech. I spent years with buildings: a PhD in architectural history, a master’s in visual communication, and four years teaching design students.

What stayed with me is how people find their way through a place, and what makes them get lost. Apps are places too. So now I design those.

Yagmur presenting at the Digital Architecture Research conference
Presenting at the Digital Architecture Research conference · Bialystok, Poland, 2025
Yagmur presenting her gamification research at the ECGBL conference
Presenting at the ECGBL Conference · Levanger, Norway, 2025

How I work

  • People first. I talk to users and test with them before and after I draw.
  • I explain my choices. Teaching made that a habit. Every screen has a reason I can say out loud.
  • Close to the team. I like working next to founders and developers, not handing files over a wall.

Where I’ve worked

  1. KonkretProduct Designer · freelance
  2. Hergele MobilityProduct Designer
  3. KitboxProduct Designer
  4. Bahçeşehir & BeykentDesign Lecturer
  5. Beykent UniversityLead Visual Designer
  6. Valencia Tech · BeykentUX Researcher

Say hello

If you’re building something people should find easy to use, I’d like to hear about it.

Email me

LinkedIn

Academia

My photos

ChengelLatest work · in App Store review

Chengel · iOS · submitted to the App Store

Learn words by solving crosswords.

Chengel is an iOS app for learning vocabulary. The clue is in your language, the answer is in the one you are learning, and you write it in my own handwriting, like filling in a newspaper crossword.

ROLE

Designer, with one iOS developer

I designed the app, the design system, the icons and the illustrations. One iOS developer built it.

PROBLEM

Word apps feel like homework

Lists and flash cards work, but they feel like studying. I wanted it to feel like a game you already know.

SOLUTION

A crossword with a garden

Solve a puzzle, keep the words you learned, and plant a flower for every puzzle.

IMPACT

Built in 3 weeks, now in review

Designed and built Oct 1–20, 2026. It is waiting for Apple’s review.

iOSDesign systemLight and dark
In App Store review

Follow the path, solve the puzzle, practice your words and grow a garden.

FOLLOW THE PATH

Every stop is a puzzle. Finish it and a flower grows.

The Chengel path in dark mode with a finished puzzle and the next one to start
SOLVE THE PUZZLE

Clues in your language. You write the answer in my handwriting.

A crossword in dark mode with the word NACHT written in handwriting
PRACTICE IT

Solved words go into three folders. Practice them so you always remember.

The word folders in dark mode: Knew it, Knew with letters and Mostly revealed
GROW A GARDEN

Each level has its own unique flowers. Each milestone helps you bloom one.

The awards screen in dark mode with the flowers from common to rare
1
Idea

I wanted learning words to feel like an old newspaper crossword.

QUESTION

How can learning vocabulary be fun, like we used to do in old times? Solving a crossword with a pencil.

IDEA

A crossword where the clue is in your language and the answer is the word you are learning.

REFERENCE

We used to solve newspaper crosswords by hand. The app brings that back, down to the handwriting. But this time it helps you learn a language.

2
Handwriting

I took the typography of my handwriting.

The letters you write are my own handwriting.

Newspaper crosswords were filled in by hand, so I wanted that look back. I turned it into a typeface, Yagmur Hand. One font file covers German, Turkish and Spanish letters.

Only the grid uses it. Everything else is a clean monospace.

Yagmur Hand: the word NACHT, the alphabet and the accented letters
The crossword grid in dark mode with NACHT written in handwritingThe crossword grid in light mode
CONCEPT

One puzzle, four steps. Then again.

1
2
3
4

Follow the path.

It shows where you are. Tap the start node and the puzzle opens.

Solve the puzzle.

Clues in your language, answers in the one you learn. Stuck? A letter costs 10 coins, a word 30.

Practice it.

Solved words go into three folders: Knew it, Knew with letters, Mostly revealed. Flash cards bring them back.

Grow a garden.

Each level, A1 to C1, has a garden with eight blooms. Learned words plant them.

3
Information architecture

First launch, then four tabs.

Chengel
First launch
  • Welcome
  • How it works
  • Never forget a word
  • Languages and level
    • I speak
    • I’m learning
    • A1 to C1
  • Daily goal
    • 1, 3 or 5 puzzles
Play
  • Level path
    • A1 to C1 sections
    • Stops and chests
  • Puzzle
    • Clue bar
    • Grid
    • Keyboard
    • Letter and word hints
  • Result
    • Stars
    • New and review words
  • Streak, reminder, new bloom
Words
  • Three folders
    • Knew it
    • Knew with letters
    • Mostly revealed
  • Word list
    • Deutsch and English
  • Flash cards
    • Show meaning
    • Again, Good, Easy
Awards
  • Progress
    • Bloom card
    • Level ladder
  • Garden
    • A1 to C1 terraces
  • Badges
Profile
  • Course and stats
  • Languages
    • I speak, I’m learning, level
  • Appearance
    • Theme: system, light, dark
    • Season theme
  • Daily goal
  • Feedback
    • Sounds, haptics
    • Daily reminder
4
User flow examples

Two flows: solving a puzzle and practising words.

FIRST LAUNCH TO A SOLVED PUZZLEFrom the first launch to a flower on the path.YesNoYesYesNoNoOpenthe appFirsttime?Welcome andhow it worksLanguages,level, goalLevel path(Play tab)Tap thestart nodePuzzle: clue,grid, keyboardWrite aletterWordright?Word locks,cells turngreenAll wordssolved?Result:stars, newwordsStuck? Hint:letter 10 coins,word 30Word shakes,“not right”Pick thenext clueStreak, reminder,new bloomBack on the path:a flower is planted
WORDS AND FLASH CARDSFrom a folder of words to a card you rate.EasilyMostlyNot yetYes: next cardNoOpen theWords tabMy words:three foldersOpen afolderWord list:Deutsch, EnglishTapStudyFlash card:the word inmy handwritingShowmeaningDid youknow it?Easy: movesup a folderGood: stayswhere it isAgain: movesdown a folderCardsleft?Back to thefoldersScreenActionDecisionStart or endSuccessProblem
5
Iterations · keyboard and dark mode

Two things I tuned: the keyboard was louder than the puzzle, and the dark colours tired the eyes.

Four keyboard options in light and dark: soft keys, tonal keys, lilac keys and letters only. Tonal keys was chosen.
Four options. I chose number 2, tonal keys.
BEFORE · DARK
Keyboard before: outlined keys, dark mode
AFTER · DARK
Keyboard after: tonal keys, dark mode
BEFORE · LIGHT
Keyboard before: outlined keys with dark letters, light mode
AFTER · LIGHT
Keyboard after: tonal keys, light mode
KEYBOARD

Keys had a 1px outline and full-ink letters, so the keyboard shouted louder than the crossword.

Now: no outline, keys one hair lighter than the panel, muted letters. Letter on key is 8.4:1 in light and 6.8:1 in dark.

DARK MODE

I tried several dark colours. Some reflected and tired my eyes.

Now: a purple-tinted night grey, never pure black.

6
Design system · colour

Three colours carry the brand: navy, purple and light blue. Purple acts, blue writes.

Chengel colour tokens in light and dark: surfaces, ink, brand, written and status colours with their values
7
Design system · type

One monospace face for the interface, and my handwriting turned into a typeface for the letters you write.

Chengel type styles in JetBrains Mono, the puzzle letter style in my handwriting, and lines, corners, space and shadows
8
Design system · components

Built with atomic design: atoms, molecules, organisms, each drawn once and reused on every screen.

Chengel components cut from the live design system, in dark mode: atoms, molecules and organisms
9
Design system · icons and illustrations

I drew the icons, Cengo and the flowers by hand, then polished them with AI.

The 52 hand-drawn icons, the spot illustrations, Cengo in seven poses and 40 blooms, in dark mode
10
Final screens · dark

Welcome, the path, a puzzle, a flash card, the garden and the profile, in dark.

Chengel dark screen 1
Chengel dark screen 2
Chengel dark screen 3
Chengel dark screen 4
Chengel dark screen 5
Chengel dark screen 6
11
Final screens · light

The same screens in light: a pale blue paper, navy ink and purple that acts.

Chengel light screen 1
Chengel light screen 2
Chengel light screen 3
Chengel light screen 4
Chengel light screen 5
Chengel light screen 6
Download from the App Store
Chengel app icon

Chengel

Learn words by solving crosswords

I sent Chengel to the App Store. It is waiting for Apple’s review and should be out soon. I will put the link here.

In App Store review
Chengel Path
Chengel Puzzle
Chengel Words
Chengel Garden
Chengel path
Prototype

Solve a puzzle yourself.

Tap the start node and type the words with your keyboard or the keys.

What it achieved, and what I'd do next.

IMPACT
Sent to the App StoreDesigned and built in 3 weeks, Oct 1–20, 2026, with one iOS developer. Now waiting for review.
One system, two looksLight and dark come from the same tokens. Every text pair stays above 4.5:1 contrast.
Something people can tryThe clickable prototype is on this page, so you can solve a puzzle yourself.
NEXT
1
Go liveAdd the App Store link here when Apple approves it.
2
Seasonal themesChristmas, New Year, Easter and Halloween are set up as colour overrides, still in design.
3
More puzzlesThe path already has scenery up to C1. It needs more puzzles to walk through.

WHAT I DID

I designed Chengel, an iOS crossword for learning words: the app, the design system, the icons and the illustrations.

HOW I DID IT

I took the typography of my own handwriting for the puzzle letters, drew the icons by hand, polished them with AI, and built the design in Claude Design.

WHY I DID IT

I wanted learning words to feel like the newspaper crosswords we used to fill in by hand.

WHO WAS INVOLVED

Me for the design, and one iOS developer who built it. It is now in App Store review.

PositionProduct Designer
Time3 weeks · Oct 1–20, 2026
ToolsClaude Design

iKeepRedesign shipped in a week

iKeep · iOS · live on the App Store

Store the things to remember.

Categorize. Find. Copy and paste them. Super fast.

A vault for the values you type again and again, like Wi-Fi passwords and door codes. I was asked to rebrand it, and in one week I redesigned how it works too.

ROLE

Rebrand and redesign

The brief was a new look. I took the app end to end, and worked with the iOS developer who built it.

PROBLEM

Every item looked the same

You had to read each row to find the one to copy, and Add sat out of thumb reach.

SOLUTION

Cards you can spot at a glance

Colour-coded cards, Add in thumb reach, categories you can reorder, sorting and dark mode.

IMPACT

Find it by colour, not by reading

The new version shipped within the week and is live on the App Store.

iOSLight and darkiCloud sync
iKeep on the App StoreApp Store ↗

Save it, find it, copy it.

YOUR CARDS

Every item is a colour-coded card. Tap a card to copy its value.

YOUR CARDS
NEW ITEM

Add from the button in thumb reach: category, title and value.

NEW ITEM
SEARCH

Search across everything you saved.

SEARCH
DARK MODE

Every screen in light and dark.

DARK MODE
1

The brief was a new look. The real problem was finding the right value.

THE BRIEF

Rebrand iKeep, an iOS vault for the values people copy again and again.

WHAT I FOUND

Every item looked the same. You had to read each row to find the one to copy, and Add sat at the top right, out of thumb reach.

WHAT I DID

I redesigned how it works too, in the same week: cards, thumb-first actions and organisation.

CONCEPT

The app in four steps.

1

Make a category.

Give it a name, an icon and a colour. Drag the chips to reorder them.

2

Save it.

Tap the plus button and add a value to a category.

3

Find it.

Spot it by colour, filter by category, search or sort.

4

Copy it.

Tap the card to copy the value. Values stay hidden until you tap the eye.

One tap away.
2
Information architecture

From Home to every action: where each screen lives.

iKeep
All items
  • Category chips
  • Item cards
    • Copy value
    • Reveal value
    • Mark as favourite
  • Favourites filter
Create new
  • New item
    • Category
    • Title
    • Value
  • New category
    • Name
    • Icon
    • Colour
Item actions
  • Swipe right
    • Edit details
  • Swipe left
    • Delete
Search
  • Search field
  • Matching items
Sort
  • Newest
  • A–Z
  • Custom order
Settings
  • Light, dark or device
  • Biometric security
  • iCloud sync
  • Language
3
User flow example · Copy a value

From the list to the value pasted in another app.

NoYesYesNoYesNoTry againCopyRevealUserBiometricsecurity on?Unlock withFace IDAll itemsKnow thecategory?Tap a categorychipType in thesearch fieldAny match?Nothing foundColour-codeditem cardsCopy orreveal?Tap CopyTap the eyeValue shownfor a momentCopiedSwitch to theother appPaste the valueDoneStartScreenActionDecisionSuccessProblem
User flow example · Add an item

From the Add button to a card that syncs.

YesNoYesNoFixUserTap Add(in thumb reach)New itemCategoryexists?Pick thecategoryNew category:name, icon, colourSave thecategoryAdd a titleAdd the valueAll fieldsfilled in?Show themissing fieldTap SaveCard isin the listiCloud syncs toyour devicesReady to copyStartScreenActionDecisionSuccessProblem
4
Before and after

What I changed in the redesign.

BEFORE

Every item looked identical.

→
AFTER

Every item is recognizable.

BEFORE

The add button sat at the top right.

→
AFTER

Add is in thumb reach. Creating an item is easy.

BEFORE

Light only.

→
AFTER

Dark mode.

BEFORE

Category filter only.

→
AFTER

Browse and sort.

5
What I added

Search, new categories, edit and delete: the features I added.

Search
New category
Delete and edit
Edit details
GET IT

Try iKeep on your iPhone.

It’s live on the App Store.

Open in the App Store ↗

What it achieved, and what I'd do next.

IMPACT
Shipped in a weekThe rebrand and the redesign went live within the week.
Find it by colourColour-coded cards replace identical rows, in light and dark mode.
NEXT
1
Test the buttonsTest where the buttons sit with real users, to check that Add and the actions are where thumbs expect them.
2
Category chips on HomePut a chip for each category next to the cards on the home screen, so filtering is one tap.

WHAT I DID

I rebranded and redesigned iKeep, an iOS vault for the values people copy again and again.

HOW I DID IT

I mapped the information architecture, turned list rows into colour-coded cards, moved Add into thumb reach and designed light and dark mode.

WHY I DID IT

The brief was a new look, but the app was hard to use: every item looked the same. A new look alone wouldn’t fix that.

WHO WAS INVOLVED

Me as the designer and an iOS developer, in one week.

PositionProduct Designer
Time2026 · one week
ToolsFigma · Claude Design

HealthBarMost complex work · research to App Store

HealthBar · iOS, Apple Watch · live on the App Store

Your health, turned into a health bar.

It started as an MVP with one idea: show Apple Health data the way an RPG shows a character’s health. The pixel-art look comes from that.

1
THE PROBLEM

Health apps hand you numbers, but not a next step.

8 of the 13 people I surveyed had muted health-app notifications that felt guilt-inducing or repetitive.

2
WHAT I DID

One score and one suggestion.

All the Health data becomes one HP bar, and Master Ronnie suggests one small thing to do next.

3
THE EVIDENCE

4 → 8 of 9testers read their HP right at first look.

I tested the first dashboard and the redesign with 9 people each.

Not measured yet: whether people check in more often once the app is live.

Me: research, UX/UI and usability tests. The iOS developer built it, and we made the product decisions together. Feb 2025 – now.
Published under the developer’s App Store account.

App Store ↗

Check your HP. Get one suggestion. Act on it.

YOUR HP

The HP bar, your day in numbers and Master Ronnie’s tip.

YOUR HP
QUICK ACTIONS

Breathe, reflect, check in your mood or log water.

QUICK ACTIONS
YOUR CHARACTER

Average stats and your HP trend over time.

YOUR CHARACTER
ACHIEVEMENTS

Achievements, daily goals and difficulty.

ACHIEVEMENTS
1

Health apps hand you numbers, but not a next step.

QUESTION

How can Apple Health data feel calm and encouraging, and tell you what to do next?

HYPOTHESIS

People will check in more often, and act on what they see, if their data is framed as a story with a mentor instead of a spreadsheet with a verdict.

WHAT THE SURVEY SHOWED

8 of the 13 people I surveyed had muted health-app notifications that felt guilt-inducing or repetitive.

2
SURVEY

I surveyed 13 people, ages 21–45, about the health apps they use.

8 of 13

have muted health-app notifications that felt guilt-inducing or repetitive

7 of 13

find charts and trend graphs hard to act on without context

9 of 13

would check a health app more often if it felt encouraging, not evaluative

7 of 13

want one clear suggestion a day, not a dashboard of metrics

6 of 13

say a widget or watch glance matters more than opening the app

“I see the line drop, I don’t know why.”

one of the people I surveyed

CONCEPT

The daily loop in five steps.

1

Check your HP.

Steps, sleep, water, mood and more become one HP score.

2

Get one suggestion.

Master Ronnie suggests one small, low-pressure thing: a short walk, one calm glass.

A mentor, not a monitor.
3

Act on it.

Breathe, reflect, check in your mood or log water in one tap.

4

Read your day.

Your day becomes a game-style combat log, with trend cards and tiny quests.

5

Level up.

Your character grows: average stats, the HP trend and achievements.

3
Information architecture

Four parts, from first run to long-term progress.

HealthBar
Onboarding
  • Meet Master Ronnie
  • Create your character
    • Name
    • Avatar
    • Daily goals
  • Connect Apple Health
  • Notifications
Daily loop
  • HP card
    • Stats
    • Master Ronnie’s tip
  • Quick actions
    • Breathe
    • Reflect
    • Mood
    • Water
  • Combat log and feed
Character
  • Stats
    • 7 days to 1 year
  • HP trend
  • Achievements
  • Goals and difficulty
  • Weekly recap
HealthBar+
  • Stress load
  • Recovery score
  • Body-age estimate
  • Pattern insights
4
User flow example · HP and the tip

From the HP bar to a tip you act on.

YesNoYesLaterYesNoUserApple Healthconnected?ConnectApple HealthHP card:bar and statsMaster Ronnie’stipDo it now?Tip stays onthe HP cardQuick actions:Breathe, Reflect,Mood, WaterDo the actionHP goes upCombat logshows itDaily goalreached?Ronnie suggeststhe next stepAchievementunlockedHP trendupdatesStartScreenActionDecisionSuccessProblem
User flow example · Quick action

From logging water or mood to an achievement.

YesNoYesNoUserHome: quickactionsWater ormood?Tap WaterChoose anamountTap MoodPick howyou feelLog itHealth accessallowed?Allow it inSettingsHP updatesCombat logand feedGoal reached?Progress onthe HP cardAchievementunlockedHP trendupdatesStartScreenActionDecisionSuccessProblem
5
Onboarding and first run

Meet Master Ronnie. The app shows its value before it asks for access.

Tap to meet Master Ronnie
Create your character
Connect the Health app
Notifications
Master Ronnie welcomes you
Tour: your HP
Tour: buffs and debuffs
6
The daily loop

The dashboard is home base. Check HP, get one suggestion, log an action.

HP bar and tip
Suggestion and widget card
Yesterday’s tale
Side quest
Quick actions
Share your HP card
7
Your character

Long-term progress: stats, trends, achievements and goals.

Level and today
Character stats
HP trend
Weekly recap
Achievements and goals
Achievement unlocked
8
Usability testing

9 testers per version compared the first dashboard with the redesign.

4 → 8 of 9

read their HP correctly at first glance

13s → 4s

average time to a correct read

3 → 9 of 9

found Master Ronnie’s suggestion without being told

2 → 7 of 9

understood the buff and debuff icons without explanation

6 → 2 of 9

asked what an RPG term meant

5 → 8 of 9

rated the screen calm to check daily

What changed: HP got one dominant colour on a quieter background, and body copy dropped a full size below the headers. The fix wasn’t more explanation. It was letting one thing lead.

9
Design system and components

A pixel-RPG palette on a dark, low-glare surface. Yellow is only for HP. Space Mono throughout.

Design system · Foundations
Design system · Components

The parts that repeat on every screen, and the ones that leave the app.

Cut from the final design.

1

Atoms

The smallest parts

App icon
App icon
Buffs and debuffs
Buffs and debuffs
Heart states
Heart states
Avatars
Avatars
2

Molecules

Atoms that work together

HP bar, empty to full
HP bar, empty to full
HP colours
HP colours
Master Ronnie
Master Ronnie
3

Organisms

Widgets and the watch

Suggestion widget
Suggestion widget
HP widget
HP widget
Small widget
Small widget
Large widget
Large widget
Apple Watch
Apple Watch
GET IT

Try HealthBar on your iPhone.

It’s live on the App Store, with widgets and an Apple Watch app.

Open in the App Store ↗

What it achieved, and what I'd do next.

IMPACT
Read at a glanceAfter the redesign, 8 of 9 testers read their HP right at first look, up from 4 of 9.
FasterThe time to a correct read dropped from 13 to 4 seconds.
LiveOn the App Store, and I keep shipping updates.
NEXT
1
Less textPeople feel overwhelmed by how much there is to read. I want the HP bar and one tip to carry each screen.
2
Keep testingTest each update with a few users before it ships, the way the dashboard redesign was tested.

WHAT I DID

I designed and shipped HealthBar with an iOS developer: an RPG-style Apple Health companion for iOS, Apple Watch and widgets.

HOW I DID IT

I surveyed 13 people, built the screens in Stitch with my own styleset, tested the first dashboard and the redesign with 9 people each, and released it.

WHY I DID IT

Most health apps give you numbers and guilt. I wanted one that tells you, kindly, what to do next.

WHO WAS INVOLVED

Me for the research, UX/UI and usability tests, and an iOS developer who built it; we made the product decisions together. 13 survey participants and 9 testers per version.

PositionProduct Designer
TimeFeb 2025 – now
ToolsStitch · Claude Design · Figma · Codex

PlayMateDeepest UX research

PlayMate · case study · community sports app

Wanna do sports but don't have a team/partner? I have a solution!

This is a mobile app that matches people for sport. Instead of a group chat, you join a game room: you see the time, the place, the cost, who's playing. And the team members? You can check their profile to see skill levels, peer ratings and a no-show penalty, and build trust beforehand.

ROLE

Solo, from interviews to prototype

I ran the interviews and the usability tests, and designed every screen.

PROBLEM

Organising a game, why pure chaos?

All 4 players I interviewed had a routine, and all struggled with coordination, skill matching or trusting someone new.

SOLUTION

Game rooms with reputations

Schedule, split the cost and play! No place left? Join the waitlist.

IMPACT

Tested at every step, 4 iterations

With the testers' feedback, I added cost splitting, waitlists and a rating flow. Made the visual style calmer.

Case studyUser researchUsability testing

Find a game. Join it. Rate your teammates.

FIND A MATCH

Filter by sport, skill and time. Each card shows spots left and the price.

GAME ROOM

Price per person, level and who's playing, before you join.

AFTER THE GAME

Did they show up? Rate each teammate on four things.

YOUR PROFILE

Skill, reliability and reviews add up to a reputation.

Playing sport regularly is easy to want and hard to organise.

QUESTION

How can I make finding and organising a game easier, and build enough trust for people to play with strangers?

HYPOTHESIS

The barrier isn't motivation. It's the effort of finding the right person, and not trusting that person or the system.

CONFIRMED

All 4 players I interviewed already had a sport routine. All of them struggled with coordination, skill matching or trusting someone new.

1
PLANNING

Before talking to anyone, I wanted to plan the research and decide how to move forward.

1
PROJECT BACKGROUND

PlayMate is a community app that connects people through sport to fight social isolation. Four hypotheses to test, and the project is in the exploration phase.

2
RESEARCH GOALS

Understand what frustrates people when they organise sport through WhatsApp or Facebook. Find which profile information makes meeting a stranger feel safe, and which features keep people playing regularly.

3
RESEARCH METHODS

Semi-structured interviews and desk research.

4
RESEARCH QUESTIONS

Seven questions the interviews had to answer, such as how people find sports partners today, what makes them trust a stranger, and what happens when a partner cancels.

5
PARTICIPANT RECRUITMENT

People who already have a sport routine, 30 minutes each, on Zoom or in person. I run the sessions myself; six were planned.

My FigJam research plan board12345678★
6
CONSTRAINTS

An app could leave out older athletes who are less comfortable with digital tools. Skill levels are self-reported and hard to verify, and new users start with no reliability rating.

7
DATA ANALYSIS

Zoom for the interviews, a phone to record, paper for notes. Data is organised in Excel and clustered in FigJam, and recordings and notes stay on my own devices.

8
DELIVERABLES AND TIMELINE

Four phases: discovery (report, personas, journey map), UX design (wireframes, information architecture), UI design (style guide, design system, final screens) and a prototype.

★
RESEARCH JAM

The anchor topics stay fixed (daily sport routine, community, social circle, motivation, trust). The questions change with each person.

2
INTERVIEWS

I interviewed 4 people who already have a sport routine, 30 minutes each, on Zoom.

METHOD

Semi-structured interviews and desk research.

FOUND

At a few gyms, in my language course and around me.

PREPARED

16 open questions, cut down to fit 30 minutes.

RAN

Consent first. On Zoom I typed notes live and recorded the audio to fill the gaps.

Four Zoom screenshots of the interviews, participants blurredThe interviews on Zoom.
WHO

BAJO · 32

iOS developer, Berlin. Basketball and ice hockey, 4× a week.

ANDREA · 34

Designer, Berlin. Jiu-jitsu, 2× a week.

ANNA · 32

Designer, Berlin. Trains alone, needs team sports to stay motivated.

MCAN · 33

Office worker, Istanbul. Gym, padel, football, tennis.

3
SYNTHESIS

Every note on one board, sorted into six themes. The yellow ones became features.

MY NOTES FROM THE 4 INTERVIEWS
1
What motivates them to play

Fun, learning, progress.

2
Who they want to play with

Strangers at my level.

3
Time and place preferences

WhatsApp group + poll.

4
Trust factors

Mindset matters more than looks.

5
Pain points

A time that suits everyone.

6
Features they wished the app had

Favourites, ratings, skill level.

My FigJam board with every interview note, sorted by hand123456
THE SIX THEMES

Numbers match my notes on the left.

Heard, not builtTurned into a feature
1

What motivates them to play

  • Meeting friends, well-being
  • Learning a new technique
  • Fun, sweat, a mood boost
  • Progress and socialising
  • A plan, so skipping is not an option
2

Who they want to play with

  • Strangers at my level×3
  • Friends×1
3

Time and place preferences

  • WhatsApp group + poll×3
  • Turn up at the gym and find someone×1
4

Trust factors

  • “Mindset matters more than looks. Ego is a red flag.”
  • Wants people her age who talk with respect
  • Reviews and ratings, like Airbnb
  • “I don’t care if they’re trustworthy”
5

Pain points

  • Finding a time that suits everyone
  • Finding a place for 12 people
  • Bringing people together
  • Finding someone of similar size and mindset
6

Features they wished the app had

  • A favourite playmates list
  • A set time every week
  • Rate people on reliability, passion, respect…
  • Skill level on the profile
  • “Why do you do sports?” on the profile
  • “At what time I’ll arrive”
  • Reminders to play
Research → features

I filtered my notes and turned them into buildable features.

WHAT IT MEANSWHAT I DESIGNED
The plan lives in a chat nobody can follow.
→

Game rooms

Time, place, player limits and cost split in one room.

Level matters more than knowing someone.
→

Skill levels

On every profile and game, and as a filter.

Trust has to be visible before the first game.
→

Peer ratings

Reliability, passion, respect and communication, after every game.

A no-show costs the people who did show up.
→

Waitlist and no-show penalty

The next person fills the spot. A no-show lowers reliability.

A good match should be easy to repeat.
→

Friends and repeat games

Favourite playmates one tap away. Repeat a game every week.

My empathy map: says, does, feels, thinks, pain and gain
4
EMPATHY MAP

What a player says, does, feels and thinks.

SAYS

Wants to play, but doesn’t know if the court is free or who will be there.

DOES

Searches Maps, Facebook and Reddit. Deletes apps that are inactive or too expensive.

FEELS

Nervous about being judged, isolated in a new city, hopeful about finding a “tribe”.

THINKS

What if I join and can’t make it? I want to know the people will actually show up.

PAIN

Flaky partners, mismatched skill levels, social anxiety, no court information.

GAIN

A consistent playmate group, a place in a new city, better health, a reliability rating that builds trust.

5

Personas & Journey map

Two kinds of player. I designed for Summer first, and checked every flow against Leo.

Personality

IntrovertExtrovert
AnalyticalCreative
BusyTime rich
MessyOrganized
IndependentTeam player
PassiveActive
SafeRisky

Bio

Moved to a new city six months ago. She sees sports as a gateway to community and long-term friendship, but finding reliable, like-minded partners feels exhausting and uncertain. The real barrier isn’t the activity, it’s the anxiety of the unknown.

Primary personaThe social player
“The sport isn’t the hard part. Meeting strangers is.”

Stands for the players who want a circle, not just a workout.

PAIN POINTS
  • Can’t tell if a stranger is respectful or there to dominate
  • Tired of looking for new people every time
NEEDS
  • Reviews and commitment ratings
  • Profiles that say why someone plays
  • A favourite playmates list
Summer

Summer

She represents the user who is motivated by the “Community” and “Loneliness” aspects of the interviews.

  • Age: 28
  • Occupation: Interior designer
  • Location: Berlin, Germany
PRIMARY MOTIVATION

To find a reliable “tribe” and turn exercise into a long-term social activity.

Personality

IntrovertExtrovert
AnalyticalCreative
BusyTime rich
MessyOrganized
IndependentTeam player
PassiveActive
SafeRisky

Bio

A busy professional who treats sports as a mood booster and reset button. He wants a game to be seamless, but planning wears him down: group chat back-and-forth, last-minute cancellations and skill-level gaps.

Secondary personaThe efficiency seeker
“If the game feels unreliable, I stay home.”

Stands for the players worn down by logistics and cancellations.

PAIN POINTS
  • Endless group chat to agree on a time
  • Last-minute cancellations and skill gaps
NEEDS
  • Scheduling from poll results
  • Reliability scores and skill filters
  • An “I’ll arrive at…” status
Leo

Leo

He represents the user who is motivated by the “Logistics” and “Reliability” aspects of the interviews.

  • Age: 35
  • Occupation: Psychologist
  • Location: Berlin, Germany
PRIMARY MOTIVATION

Efficiency, consistency, and a “no-drama” experience.

MY JOURNEY BOARD IN FIGJAM

One colour per phase. Rows: goals, actions, feelings, opportunities.

PlanningCoordinationThe matchAfter
My FigJam journey map board1234
JOURNEY MAP

Summer’s game, step by step: where it hurts, and what the app does there.

GOAL
Find a reliable game and partner
Organise or join with little effort
A safe game at the right level
Build a circle, plan the next one
DOES
Picks a sport, looks for a game and a place
Checks who’s coming, books the court, reposts if short
Shows up and checks in
Rates partners, adds favourites
FEELS
Hopeful, but anxious about who she’ll meet
Annoyed by group chat chaos
Relieved it started; curious about the others
Satisfied, but too many rating fields
OPPORTUNITY
Show mindset and level before meeting
Automate polls and arrival updates
A check-in that counts towards reliability
Favourite playmates to build a circle
IN THE APP
Skill level · “I do sports because”
Game room · player limits · waitlist
Attended? yes / no · no-show penalty
Four ratings · friends
CONCEPT

The app in three steps.

1

Set up a profile.

Pick your sports, set your skill level and say how you like to play.

Your profile is your reputation.
2

Find or create a game.

Browse open game rooms near you by sport, level, time and cost. Join in one tap, or start your own.

Scheduling, cost split and waitlist are handled for you.
3

Play, rate, repeat.

After the game, rate your teammates on reliability, passion, respect and communication.

The more reliably you show up, the easier it is to get into games.
6
Information architecture

Five tabs. Create sits in the middle, one tap from anywhere.

Open the app→4 intro screens→Sign up, log in or guest
PlayMate
Home
  • Upcoming matches
    • Game details
      • Photo and details
      • Participants
      • Join
      • Add to calendar
  • Quick links
    • Match history
    • My ratings
  • Games to rate
    • Rate the game
  • Explore nearby
Discover
  • List or map
    • Game
      • Game details
  • Filter: sport, skill, time
  • Closest match preview
Create
  • Sport
  • Location
  • Date, time, players
  • Skill filter
  • Create the game
Friends
  • Recent chats
  • Friend list
    • Friend profile
      • Add friend
      • Message
      • Invite to game
  • Post-game feedback
Profile
  • About
  • Reliability rating
  • Reviews
  • No-show log
  • Settings
7
User flow example · Join a game

From finding a game to rating the people you played with.

YesNoYesNoYesNoUserSigned in?Sign up, log inor guestDiscoverFilter by sport,skill and timeList or mapof gamesGame details:photo, playersSpots left?Join the gameSee your shareof the costYou are inAdd to calendarUpcomingmatchesJoin thewaitlistSpot opens?Find anothergamePlay the gameRate yourteammatesReliabilityrating updatesStartScreenActionDecisionSuccessProblem
User flow example · Create a game

From the middle tab to a game other people can join.

YesNoNoFixYesUserTap Create inthe tab barChoose a sportPick the courtor venueDate, time andnumber of playersSet a skilllevel filterSet the cost,split per playerAllowwaitlist?Extra playersjoin a queueReview thegameAll fieldsfilled in?Show whatis missingCreate the gameGame is liveShows in Discoverand HomeInvite a friendto the gamePlayers join,or wait in the queuePost-gamefeedbackStartScreenActionDecisionSuccessProblem
8
Low fidelity wireframes

I tested the flow in grey wireframes before any styling.

1

Discover: Filter by sport, skill and time, then pick a game from the list or the map.

2

Game room: Where, when and the average skill level, then who has joined. Join, or add it to the calendar.

3

Create a game: Sport, place, date, player limits and skill level in one form.

4

Profile: Skill level, preferred sports, “I do sports because”, ratings and reviews.

THE WHOLE BOARD
My grey wireframes board1234
1
2
3
4
9
Usability testing

4 testers, moderated, 30 minutes each, on every version. What each round changed:

WWireframes
Cost split, venue booking, player limits and a waitlist (shown above), plus:
Rating flowNo clear moment to rate. A rating screen now opens when a game ends.
No-show penaltyA confirmed player who doesn't come loses a reliability star.
1Hi-fi 1
ContrastCards blended into the background. Raised.
MapStays fixed while the list scrolls. Pins redrawn.
SettingsLost its tab. Opens from the profile menu.
OnboardingFirst-time users had none. Designed and added.
2Hi-fi 2
Guest mode"Continue as guest" added to onboarding.
CreateOpens as an overlay, not a separate screen.
Past gamesGreyscale, so they read as not editable.
3Hi-fi 3
OnboardingCopy shortened. Progress dots to go back.
"Welcome"Repeated on almost every screen. Removed.
InputsEmpty fields looked focused. Now clearly different.
Testing round 1 · Game room

Testers asked what a game costs and where to book the court. Now both are on the page.

Game room before and after testing round 1
Testing round 1 · Create a game

Testers asked how many players a game needs, and what happens when someone drops out.

Create a game before and after testing round 1
10
Iterations

I made 4 iterations. To compare them clearly, I show the first and the final version of the four screens that changed most.

Map
HI-FI 1
→
FINAL
No-show penalty
HI-FI 1
→
FINAL
Profile
HI-FI 1
→
FINAL
Create a game
HI-FI 1
→
FINAL
11
Design system and components

Why it looks like this: vibrant neo-brutalism. Six colours, each with one job, and two typefaces.

Design system · Foundations board
Design system · Components

Built with atomic design: atoms form molecules, molecules form the sections of every screen.

Exported from the final Figma file.

1

Atoms

The smallest parts

Filter chips
Filter chips
Segmented switch
Segmented switch
Primary button
Primary button
Text input
Text input
Search field
Search field
Interest tags
Interest tags
Close button
Close button
2

Molecules

Atoms that work together

Quick link
Quick link
Player row
Player row
Skill rating
Skill rating
Past game (greyscale)
Past game (greyscale)
Player settings
Player settings
Schedule
Schedule
3

Organisms

Sections of a screen

Upcoming match
Upcoming match
Game card
Game card
Price and details
Price and details
Map preview card
Map preview card
Court card
Court card
Incident log
Incident log
Starting lineup
Starting lineup
Trophy case
Trophy case
Teammate rating
Teammate rating
Reliability impact
Reliability impact
Performance metrics
Performance metrics
Tab bar
Tab bar
Cost management
Cost management
Venue location
Venue location
Profile menu
Profile menu
12
Final screens

Onboarding, home, the map, create a game, friends, a friend's profile and the no-show penalty.

PlayMate screen
TRY IT

Play with the prototype.

Join a game, create one and rate your teammates, on the final screens.

Open the prototype ↗

What it achieved, and what I'd do next.

IMPACT

There are no live users yet, so no conversion numbers. What the work did show:

The problem is realAll 4 interviewees struggled with coordination, skill matching or trust.
The flows workEach test round removed what testers stumbled on, from missing payments to tiring colours.
NEXT
1
PilotRun it with real players to measure show-up and retention.
2
Vibe filterCompetitive, casual or fitness. Mindset mattered as much as level.
3
ComponentsOne version of each toggle, star rating, shadow and chip.
4
Long listsTabs or a carousel, so past games don’t fall off screen.

WHAT I DID

I designed PlayMate end to end: a community app that matches people for sport through game rooms instead of group chats.

HOW I DID IT

I planned the research, interviewed 4 people, sorted the notes into themes and turned them into features. Then I tested the wireframes and each hi-fi version.

WHY I DID IT

You want to do sport regularly. Deciding is easy, but the hard part is finding a partner, trusting them and finally organising the game. I wanted to solve that problem.

WHO WAS INVOLVED

It is a bootcamp project. 10 mentors, 5 classmates, 4 interviewees, 4 testers and me.

PositionProduct Designer
Time3 months · Summer 2026
ToolsFigma · Claude · Stitch