Safina is a mobile app that helps people manage their finances in a
more natural and intuitive way.
With its conversational AI and clean interface, Safina transforms
expense tracking into a simple, human experience allowing users to
record spending or income just by talking to the app.
Many People Struggle to Stay Consistent in Managing Their Finances
Most people start using finance apps with good intentions, but
after a few weeks, they stop. Manually entering every expense
feels repetitive, and the process quickly becomes a chore.
Safina was created to change that habit. This mobile app helps
users track their spending and income through simple
conversations with an AI chatbot. By saying things like
βI spent Rp25.000 on coffee this morning,β
the app automatically records and categorizes the
transaction.
As a UI/UX Designer, my role was to refine the user flow, design
the interface, and ensure that the concept felt simple, natural,
and engaging for everyday use.
Understanding The Problem
Most finance apps focus on numbers and forms, not on how people
naturally interact. Users often lose interest because the process
of tracking feels like extra work.
Safinaβs goal is to
make financial management feel effortless. Instead of navigating
through multiple forms and buttons, users can communicate directly
with the app using everyday language. The challenge was to design
this conversational experience while keeping it functional, clear,
and enjoyable to use.
Research And Direction
Although I did not conduct formal surveys or usability testing, I
studied several popular finance apps to understand where users
typically drop off. Many apps make tracking too manual or too
complex.
From those insights, I focused on creating an experience that
reduces friction, simplifies decision-making, and builds comfort
through clarity. This direction guided how I approached the user
journey, layout, and visual hierarchy.
Structuring The Experience
When I joined the project, the concept and early flow already
existed, but it needed refinement. I worked on clarifying how
users would move through the app, from onboarding to chatting with
the AI and reviewing their financial summaries.
Each step was designed to feel continuous, as if the user is in
one ongoing conversation rather than switching between different
screens. This made the experience smoother and helped maintain a
natural rhythm in how users interact with their finances.
Wireframes And Layout Exploration
After refining the structure, I translated it into wireframes to
define how each screen connects. The goal was to keep everything
minimal and functional while highlighting the core interaction:
the conversation.
I experimented with different layouts to make sure the chat
interface remained the focal point, with quick access to insights,
summaries, and recent transactions. This stage allowed me to
confirm the balance between information and simplicity before
moving to high-fidelity design.
Visual Design System
Finance can often feel overwhelming, so I aimed to make the
interface look calm and approachable. I used a combination of
soft colors, rounded shapes, and clean typography to build trust
and focus. Spacing and hierarchy were carefully adjusted to make
information easy to scan and understand.
The overall visual direction was designed to give users a sense
of comfort and control, encouraging them to track their finances
regularly.
Final Design Overview
The final design combines a conversational interface with a
structured financial overview.
Key highlights include:
π Conversational input to record expenses or income naturally
π A simple dashboard that visualizes spending patterns
π Smart categorization powered by AI
π A calm and clean interface that supports consistency
This approach makes finance management easier, transforming a
routine task into a friendly daily habit.
Designing For AI Conversation
One of the key parts of this project was defining how the app
communicates with users.
I worked on creating conversation patterns that feel natural but
remain clear and functional.
For example: βKopi 25rbβ β βTransaksi #1: Kopi β Rp 25.000 Ada
yang salah?β
By designing these interactions carefully, the chatbot feels both
capable and friendly, guiding users without confusion.
Future Steps
Safina still has room to grow.
The next planned features include:
π Bill scanning to convert receipts into data automatically
π Goal tracking to help users plan savings
π AI-driven financial insights to give personalized advice
These ideas aim to make Safina more powerful and proactive in supporting usersβ financial goals.
Reflection
Safina was a meaningful opportunity to explore conversational design in the context of personal finance.
It challenged me to think about how AI can make daily financial management easier and more natural.
My main contributions included:
π Refining and structuring the user flow
π Designing a conversational interface
π Building a consistent and calm visual design system
Through this project, I learned how to connect technology, behavior, and emotion in a way that feels seamless for users.