Case study, 2026

Payungku

A savings app that tells you how much is safe to save this week, and shows its working.

12 of 24 days covered. Rent, bills and food, not a percentage.

Figma Canva Mobbin Pinterest ChatGPT Claude Excel
Why RM12? The balance, rent, phone bill and usual spending, taken away line by line to RM37.
Home. Total savings RM900, covered for 12 days, and one card: You can save RM12 this week.
RM12 saved. Covered for 12 days.
RoleProduct and UI/UX designer
TeamSolo concept project
Scope120+ screens, 28 exception states
Tested with21 people

The problem

People want a safety net. Most never build one.

What the surveys say

63%say an emergency fund is their main reason to save2021
71%save RM500 or less a month2023
67%have savings that would last under three months2023

What I make of it

What I think is going on

In the middle of a month, it's hard to tell whether RM50 is spare or already spoken for. So people wait until payday, and by then it's gone.

The problem

People who want an emergency fund can't tell, week to week, how much is safe to put aside. So they save irregularly, or not at all.

The question

How might we tell someone what's safe to save this week, and show why?

And 42% of young earners said they'd save more with a plan built around their goals.

RinggitPlus Malaysian Financial Literacy Survey, 2021 and 2023. PIDM and iMoney survey of 1,525 Malaysians aged 24 to 35, 2019.

How Payungku was made

Process

  1. 01

    Discover

    16 apps

    136 journeys recorded, and two national surveys.
  2. 02

    Define

    3 people

    One weekly decision, and an amount that is safe.
  3. 03

    Design

    120+ screens

    The sum, the screens, and a system another designer can pick up.
  4. 04

    Prototype

    4 ways in

    Real behaviour: a PIN that fills as you tap, a code that arrives as a message.
  5. 05

    Test

    2.19 / 3

    21 people used it, then scored it. Rated Excellent.

Next: time the sign-up and the save, and see where people stop.

Research

What exists, and what is missing.

    • Cleo
    • Oportun

    This is where Payungku starts.

As recorded in my benchmark of 16 apps, 2026.

What I sawScheduled saving used a fixed amount, whatever the week looked like.

What Payungku doesEvery plan is skipped on a low week.

What I sawA secure approval for every single action.

What Payungku doesA PIN only when money leaves.

What I sawThe longest setup ran to seven or more screens.

What Payungku doesReady-made plans: one tap, then Create.

136 journeys recorded across 16 apps, 11 features compared.

The gap

People are asked to trust a number they can't check.

User personas

Three people the app was designed around.

Hidayah set the direction. Aina and Azman stand for the two groups I tested with.

  1. Illustration of Hidayah

    Hidayah Primary Early in her career

    "I'd save more if I knew it wouldn't leave me short before payday."

    • Paid RM2,200 on the 28th
    • Rent RM650
    • Saves when she remembers

    Needs To know an amount is safe before she commits.

    So the app Suggests one safe amount a week, and shows the sum behind it.

  2. Aina Final-year student, 21

    "I want to save, but I never know what I'll have next month."

    • About RM500 a month
    • No fixed payday
    • Tutors part-time

    Needs To be told when not to save, without feeling she failed.

    So the app Says "Nothing to save this week, and that's okay", and makes Skip one tap.

    Not solved yet Payungku works from payday, and she doesn't have one.

    Illustration of Aina
  3. Illustration of Azman

    Azman Administrative officer and father of three, 45

    "Show me exactly what you're doing with my account before I agree to anything."

    • Paid RM4,800 on the 25th
    • Loan, rent and school costs
    • Careful with anything new

    Needs To see what is shared, and to stop it at any time.

    So the app Lists what it will and won't do before linking a bank, and never moves money without his PIN.

Personas built from desk research and refined after testing with 21 people.

Structure

Three journeys. One of them is the product.

Once 8 stages

Getting in

From welcome to the first Payung. A person does this a single time.

  1. Welcome
  2. Phone number and code
  3. MyKad
  4. Selfie
  5. Check your details
  6. PIN and Face ID
  7. Link bank
  8. First Payung

The first saving unlocks 12 hours later. A safety wait for every new sign-up.

Every week 4 steps

The save

One amount, the reason for it, then a yes. This is the product.

  1. See the safe amount, on Home

    Optional Why RM12? The full sum, one tap away.

    If no amount is safe “Nothing to save this week. Rent took most of it, and that's okay. We'll look again on Monday.”

  2. Confirm
  3. PIN
  4. Saved

Now and then 5 actions

Looking after a Payung

What people do between saves.

  • Add money
  • Withdraw
  • Move to another Payung
  • Rename or change how it saves
  • Close

Withdrawing is instant and is never shown in red. It is their money.

Everything else in the app exists to make the middle one trustworthy.

Four tabs

Home

What should I do this week?

Payungku

What am I saving for?

Insights

What is coming before payday?

Profile

What does the app know about me?

Ask Payu

Why, and what if? Opened from Home.

Wireframes

What each screen has to do, before how it looks.

Key screens in low fidelity, showing structure and priority without styling.

Low-fidelity wireframes of ten Payungku screens: Welcome, Link your bank, Home, Why this amount, Confirm then PIN, Saved, Payungku, Emergency, Ask Payu, and When it fails.
Grey bar, text. Crossed bar, a picture. Red dot, a note. On a phone, slide the board sideways. Click it to read the notes.

Interface

See the number. Check the sum. Save.

The Payungku welcome screen. Save a little, when it's safe.
WelcomeThe brand, then one promise.
Home. Total savings RM900, covered for 12 days, and one card: You can save RM12 this week.
HomeOne primary action.
Why RM12? The balance, rent, phone bill and usual spending, taken away line by line to RM37, the most that is safe.
Why RM12?The working, one tap away.
RM12 saved. Covered for 12 days.
SavedWhat it means, in days.
The Payungku tab, listing Emergency, Hajj and Holiday, then ready-made plans.
PayungkuYour goals first. Plans below.
The Emergency Payung: covered for 12 days, RM500 of RM1,000.
EmergencyCounted in days covered.
Ask Payu answering whether RM200 is comfortable this weekend, with the working shown.
Ask PayuAnswers from your numbers.
Insights: the balance drawn from today to payday, with the bills that are due.
InsightsBalance to payday.

Design system

Colour is kept for meaning.

Built on Material 3, with every token re-pointed to Payungku. Change one value and every screen follows.

Navy#1E2A5EIndigo#4650B8Lilac#C4B5FDSky#93C5FDMagenta#EFA8DDPage#F4F4F7
IndigoTap me, or your choiceGreenIt workedAmberTake careRedFailed, or cannot be undoneBlueWorth knowing
Aa

Poppins

Titles, card names and large amounts.

RM12 saved

Aa

Inter

Every list row, label and button.

Rent due Thursday · RM650

Save RM12SkipMove moneyWhy RM12 this week?RM12 saved this weekSharing ends in 7 days
127interface icons
26goal icons
8goal illustrations
5moods for Payu, the mascot

Edge cases and error states

Errors say what happened, and what is still safe.

28moments where it can go wrong. Each one designed, so nothing is left for engineering to guess.

That didn't go through. Nothing was moved. Your RM12 is still in your Bank Islam account. Try again. 1 2 3
  1. 1
    What happened

    The app takes the blame. “We couldn't read that”, not “Invalid image”.

  2. 2
    What happened to your money

    Say what was not lost. “Nothing was moved.” “Nothing was shared.”

  3. 3
    One way forward

    One main button. Never a dead end.

All 28, by journey

Failed Try again Worth knowing

The face, the badge and the words all say the same thing.

Signing up

  • Number already registered
  • Wrong code
  • Code expired
  • MyKad unreadable
  • Face not matched
  • Identity being checked
  • Under 18
  • PINs don't match
  • Bank link not completed
  • Offline
  • Camera access denied

Logging in

  • Wrong PIN
  • Locked out
  • Face ID not recognised
  • Session timed out

Saving

  • Transfer failed
  • Nothing safe to save
  • Balance changed
  • Wrong PIN

Everyday

  • Bank connection lost
  • Sharing about to end
  • Sharing ended
  • Adding above what's safe
  • Withdrawing too much
  • No search results

Ask Payu

  • Amount not safe
  • Question it can't answer
  • Can't reach your account

User testing

Excellent. Everyone found it supportive.

  • 2.19out of 3, overall
  • 21 of 21rated it supportive
  • Nonegave a negative answer

How it was measured

1 · Used the prototype

21 people tried Payungku.

2 · Rated eight word pairs

Each person picked a point between two opposite words.

3 · Compared with other products

The scores are placed against a published benchmark.

Short User Experience Questionnaire (UEQ-S). Answers are converted to a score from −3 to +3.

It scored in the top band.

Better than 90% of the products in the benchmark.

2.20Does it work for me?Pragmatic quality · Excellent
2.18Do I enjoy using it?Hedonic quality · Excellent

With 21 people, the true score could sit anywhere from 1.98 to 2.40. All of that range is still Excellent.

Everyone found it supportive. Clarity is next.

21 of 21Supportive
14 of 21Clear

People who rated it 6 or 7 out of 7.

Supportive is the point of the app. Clear is the next thing to fix.

Working adults found it harder, but newer.

18 to 23 42 to 47
Harder
Easy2.231.88
Exciting2.311.88
Newer
Leading edge2.232.38
Interesting2.152.25

An unfamiliar idea that needs a clearer explanation.

Overall: students 2.24, working adults 2.11. Two small groups, so this is a signal, not proof.

What testers told me

Feedback from the 21 sessions, paraphrased. Hover a bubble to see who raised it and what I'd change.

Payu

How far to trust it

Twenty-one people in two small groups, so the age gap is a signal, not proof. The questionnaire measures how the app felt, not whether tasks were finished. The hedonic answers were less consistent (alpha 0.56) than the pragmatic ones (0.74). Timed tasks are the next test.

13 students and fresh graduates aged 18 to 23, and 8 working adults aged 42 to 47. Eight word pairs, each on a seven-point scale.

Pragmatic quality (usable)

2.20

Hedonic quality (engaging)

2.18

Overall

2.19
  1. −1
  2. 0
  3. 1
  4. 2
  5. 3

Zones from the UEQ-S benchmark. Brackets show the 95% confidence interval. The full scale runs from −3 to +3. Nobody scored below 0.

The eight scores.

  • Pragmatic quality (usable)2.20
  • Supportive+2.57
  • Easy+2.10
  • Efficient+2.24
  • Clear+1.90
  • Hedonic quality (engaging)2.18
  • Exciting+2.14
  • Interesting+2.19
  • Inventive+2.10
  • Leading edge+2.29
  • 0+3

Where the ages part.

Adults lowerAdults higher

  1. Supportive−0.12
  2. Easy−0.35
  3. Efficient+0.02
  4. Clear−0.25
  5. Exciting−0.43
  6. Interesting+0.10
  7. Inventive−0.15
  8. Leading edge+0.15