ClientPay

Project overview

ClientPay is a mobile app for iPhone and iPad that helps freelancers create invoices, track payments and follow up with clients without heavy accounting software. As the founder, designer and engineer, I built the first version in a six-week sprint (August-September 2025). Since launch I have shipped incremental updates based on real user feedback.

I wrote this to show how I reason through product decisions, not only what ended up in the app.

Personal and problem context

As a freelance developer and designer juggling multiple clients, I constantly struggled to produce professional invoices and remember who owed me what. Spreadsheets were time-consuming and easy to misplace. Full accounting suites felt like overkill for a handful of monthly invoices, so overdue payments often went ignored.

This frustration became the seed of ClientPay. I wanted an app that:

• Allowed me to create invoices quickly and look professional, ideally in under a minute when things go smoothly.
• Gave me instant clarity on payment status, highlighting which invoices were paid, pending or overdue.
• Automated follow-ups without sounding robotic, using AI to draft friendly chase messages.
• Supported multiple business identities and 45+ currencies so I could invoice different clients in different currencies.
• Worked offline and stored data locally so I remained in control, with optional server integration for chase messages.

The friction I experienced as a freelancer, combined with the lack of a dedicated tool for simple yet flexible invoicing, motivated me to design and build ClientPay.

Users

Primary

Busy freelancers and small teams who bill clients on a project or hourly basis. They juggle multiple clients, may operate under different business names and need to invoice in various currencies. They are comfortable with mobile apps but do not want to learn full accounting software.

Secondary

Small agencies or contract teams managing a handful of clients. They need the same features as freelancers but want recurring invoices and easy multi-business switching too.

Not targeted

• Large enterprises or firms requiring complex financial integrations and detailed reporting.
• Users needing payroll, tax calculations or full double-entry accounting.
• Anyone without access to a smartphone.

Goals & success metrics

I set concrete goals so the work stayed tied to measurable outcomes.

Constraints & assumptions

• Solo founder: With no team, I balanced scope with feasibility and focused on features delivering the highest impact.
• Local-first design: To avoid backend complexity and respect privacy, all data lives in AsyncStorage. Users can optionally wire their own API to generate chase messages.
• Free tier limitations: The free version allows creating only one invoice; additional invoices require a Pro subscription. This constraint kept the free tier simple and drove conversion without compromising the value proposition.
• Deferred payment integration: Direct payment processing (e.g., credit card links) and accounting integrations were out of scope for v1. Instead, invoices can be exported as PDFs and sent manually.
• Assumed AI acceptability: I assumed users would appreciate AI-generated follow-ups if they could edit messages. To mitigate risk, the app falls back to a static template when the AI fails.

Competitive analysis

Design process

Final screens

A guided tour of the shipped app, with callouts on the design decisions behind each screen.

1

1

2

3

4

Screen 1

Home — Dashboard

A calm snapshot of earnings and outstanding work.

• 1
  Personal greeting

  Sets a friendly tone and orients the user to their own business.
• 2
  Paid vs pending

  Payment visibility principle — the two numbers freelancers care about most.
• 3
  Recent activity

  Quick re-entry point into any in-flight invoice without searching.
• 4
  Primary CTA

  Create Invoice is always one tap away — the fast-first principle.

2

1

2

3

Screen 2

Clients

A lightweight CRM built for people who hate CRMs.

• 1
  Search + add

  Fastest possible path to either finding or adding a client.
• 2
  Avatar + name

  Human-first list design — you recognise the person before the data.
• 3
  Status at a glance

  Outstanding balance surfaced inline so you never dig to find it.

3

1

2

3

Screen 3

Client detail

Everything about a client in a single scannable card.

• 1
  Contact header

  Name, email, phone grouped so the user can act in one glance.
• 2
  Lifetime stats

  Total billed + paid builds trust and context before sending a new invoice.
• 3
  Invoice history

  Chronological list keeps past work close without leaving the profile.

4

1

2

3

Screen 4

Invoices

Status-first list that answers "who owes me?" in seconds.

• 1
  Filter tabs

  Paid / Pending / Overdue — the only three states that matter.
• 2
  Coloured status pill

  Visual status beats reading — green, amber, red map to user intuition.
• 3
  Amount prominent

  The number you care about is the largest element in the row.

5

1

2

3

Screen 5

Invoice detail

All the context needed to send, track or chase an invoice.

• 1
  Invoice header

  Number, client and total — the three facts users confirm before any action.
• 2
  Line items

  Clean breakdown with per-item totals so nothing feels hidden.
• 3
  Action buttons

  Send, share and chase live at the bottom where the thumb naturally rests.

6

1

2

3

Screen 6

Reminders

AI-drafted chase messages that still feel human.

• 1
  Single Reminders tab

  One place for every overdue invoice — addressed a direct user testing complaint.
• 2
  Editable AI draft

  OpenAI generates the first pass, but the user always owns the final message.
• 3
  Send with confidence

  Preview + send keeps the automated yet human principle intact.

Information architecture

Navigation stays shallow with a floating bottom tab bar and four core tabs: Dashboard, Clients, Invoices and Reminders. Settings and Currency are accessible from any tab via modal overlays. First-time users complete a short onboarding flow before reaching the dashboard.

Each tab serves a distinct purpose.

Core actions sit close to the surface, without burying people in nested navigation.

User journey flow