Copy Cat · Case study
Thanks for Visiting
Reach out
Context
The short version · ~10 min read
Your Mac remembers one copy. Can you see the one before it?
The problem was never remembering. Every clipboard manager does that. It is recognising: what comes back is a list of grey rows that all look the same.
My process went from my own need and desk research to 4 decisions, starting with not writing the engine, and ended on making every copy a card. The output is a strip you read by sight, each card carrying the app it came from, what it is and a preview of the thing itself, live at 1.0.0 with everything staying on your Mac. The outcome is that people pay for it, bought once rather than rented after 7 days free.
- 21+
- Apps Copy Cat recognises, each with its own colour and icon
- 4
- Card types, each with its own badge and preview
- 24 days
- First commit to the launch build

02
Background and process
From Fork to Shipped
Copy Cat is a fork of Maccy, an open-source clipboard manager that already worked and that nobody enjoyed looking at. I kept the engine and rebuilt everything above it: the interface in Figma, the trailer in After Effects, the site in Next.js, checkout on Lemon Squeezy, updates over a signed Sparkle feed.
Forked from Maccy by p0deje, MIT licensed. The engine, shortcuts, updater and settings are its work. Everything you can see is mine.
The process, in 15 steps
Open the at the top of the screen to jump between steps at any time.
Problem
Research, insights and personas
03
Problem statement
Remembering is not the broken part. Every clipboard manager already does that.
Recognising is. What comes back is a list of grey rows that all look the same, so finding the one you copied earlier means reading the list line by line instead of seeing it.
04
Research
My own reason first, then desk research: how much people paste, how they find things again, and what a clipboard holds.
Built for 1
The clipboard tool I wanted already existed, but it was paid. So I built my own, with my own touches.
My own need
11%
Paste is the most used command in Word 2003, over 11% of all commands. In Excel and PowerPoint it leads too.
Recognition over recall
Raluca Budiu, Nielsen Norman Group
“Recognition is easier than recall because it involves more cues.”
Clipboards get read
In 2020, researchers caught popular iPhone apps, TikTok among them, reading the clipboard every time they opened.
- Passwords
- Account numbers
- Shopping lists
05
Key insights
01
Paste is the habit
People paste more than they do anything else in Office, yet the Mac keeps only the last copy.
02
Seeing beats reading
Recognising is easier than recalling. A list of grey text rows gives nothing to recognise.
03
The clipboard is private
It holds passwords and account numbers, and apps have been caught reading it. Keeping every copy raises the stakes.
04
Mine to shape
The tool I wanted was paid. Building my own meant every detail could fit the way I work.
06
Personas
Proto-personas, built from the research and from who Copy Cat's features serve. Composites, not real people.
- Hex codes
- Screenshots
- Goal
- Paste the accent colour she copied this morning without opening the file again.
- Frustration
- Every copy since has pushed it out of the clipboard.
Copy Cat for themSearch across every copy, and click to paste
- Snippets
- API keys
- Goal
- Reuse yesterday's snippet without digging back through tabs.
- Frustration
- Secret keys end up sitting in history with everything else.
Copy Cat for themSensitive copies flagged, and apps switched off
- Links
- Quotes
- Goal
- Keep an essay's sources together until it is handed in.
- Frustration
- History scrolls away, and a link in a list is a long URL.
Copy Cat for themLink thumbnails, pinboards and PDF export
07
How might we
How might we let people see and reuse anything they copied, at a glance, without leaving what they are doing or giving up their privacy?
Process
Requirements to iteration
08
Design requirements
Each insight became something the design had to do. The 4 decisions further down are how each one got met.
09
IA and user flow
Where everything lives first: Copy Cat's 3 surfaces, what sits in each, and the 1 history they all read. Then the 2 flows that matter: the first run, and Nia pasting this morning's colour.
Information architecture
- Start
- Action or screen
- Decision
- Done
- Loop back
First run
Nia's paste
10
Design decisions
The engine for this already existed and was free. Every decision below is about what to do with the 3 hours a day it saved.
Whether to write the engine at all
Not taken
Build the clipboard engine myself
Shipped
Fork Maccy, rebuild what you see
How I got there
Maccy already worked and nobody enjoyed looking at it. The part that was missing was never the engine, so writing a second one would have been months spent arriving back at the start line.
What it cost
- User
- The shortcuts and settings are Maccy's conventions, not mine.
- Technical
- Every new idea has to fit an architecture I did not choose.
- Business
- Named as a fork forever, on a product people pay for.
What comes back when you look
Not taken
A faster list of text rows
Shipped
A card with source, type and preview
How I got there
Every clipboard manager fixes the remembering and stops there. Recognising the one you want is the part nobody built, and you cannot recognise a row of grey text.
What it cost
- User
- Fewer items on screen at once than a list would give you.
- Technical
- 21 app designs, 4 card types, and links that fetch their own thumbnail.
- Business
- The product now rests on how it looks, not on what it stores.
Where the icons come from
Not taken
Draw a set of my own
Shipped
Apple's SF Symbols throughout
How I got there
A Mac app that draws its own arrows looks foreign on the platform it lives in. This one sits on top of whatever you are already doing, so looking foreign is the one thing it cannot afford.
What it cost
- User
- Every glyph is one they already know from every other Mac app.
- Technical
- Tied to Apple's set, so a missing symbol has no fallback I control.
- Business
- No icon language of its own, so the brand sits entirely on the green.
How it gets paid for
Not taken
Rented monthly
Shipped
Bought once, after 7 days free
How I got there
It lives in the menu bar and does one job. The trial takes no card and no account for the same reason the price is not a subscription: neither is a thing this app has earned the right to ask for.
What it cost
- User
- You buy a version, not a promise that the next one is included.
- Technical
- Updates ship over a signed Sparkle appcast, with no subscription to check.
- Business
- No recurring revenue. Version 2 gets paid for by new buyers or not at all.
11
Evaluation
Tested against 10 heuristics
Before a usability study, a review against Nielsen's 10 usability heuristics.
8 met · 2 partly
Visibility of system status
Live permission status at setup, and a trial pill in the strip.
Match with the real world
Each card names the app it came from and what it is.
User control and freedom
Skip for now at setup; an excluded app is hidden, never deleted.
Consistency and standards
SF Pro and SF Symbols, so it reads as part of macOS.
Error prevention
Clearing history and deleting a board both ask first. Pins survive.
Recognition over recall
A preview of every copy instead of a row of text.
Flexibility and efficiency
Quick Paste on ⌥A and ⌥D, pin with ⌘L, your own shortcuts.
Minimalist design
The strip is quiet, but the menu bar menu runs to 14 items.
Recovering from errors
Empty searches suggest nothing; the permission alert asks for a restart.
Help and documentation
7 setup cards, open again anytime from How to Use Copy Cat.
12
Iteration
What broke, and what changed
From the fork to launch, 13 May to 6 June, from the git history.
Output
The product and its system
13
The final product
See what you copied.
A card carries the app it came from, a badge for what it is, and a preview of the thing itself. Links go and fetch their own thumbnail and title. The result is a strip you read by sight, at the speed you read a shelf, rather than a list you read a line at a time.
Interactive Product Experience
A sample, not the Mac app. It cannot see your real clipboard. Refresh to reset
What else it does
History scrolls away, so anything you need twice goes on a named board with its own colour. Drag a card onto the tab, and pin the ones you reach for daily.
14
Keeping its promises
A tool that sees everything you copy has to keep its own promises.
Private by design
Your history stays on your Mac, with no account and no cloud. Secrets get a shield, and any app can be switched off.
Built on its own system
1 green, 4 card types, and Apple's own type and icons, so it reads as part of the Mac rather than a guest on it.
Design System
A basic preview of Copy Cat's design system: the colours, type, card types and logo it is built on.
Copy Cat sits on top of whatever you are already doing, in a panel that covers your work for a second and then gets out of the way. So the interface is dark, quiet, and built out of one green. Everything else in the strip is colour that belongs to the respective apps.
Brand colours
Typography
SF Pro
ABCDEFGHIJKLMNOPQRSTUVWXYZ
abcdefghijklmnopqrstuvwxyz
0123456789
The app types in the system face at system sizes, because a Mac utility that ships its own font announces itself as a stranger. The site is the opposite problem and gets its own: Satoshi for everything, Instrument Serif where it needs a voice.
Card types
Everything you copy lands as one of four cards: IMAGE keeps the picture, TEXT the words, LINK the page's thumbnail and title, FILE the document and its size. Each type owns a colour: cream, teal, emerald, steel. The header wears it when the app it came from is unknown.

Logo making process

Session 1
The name was already settled, so the cat was the only open question.
Outcome
What changed, and what is next
It shipped, and people pay for it.
Copy Cat is live at version 1.0.0 for macOS 14 and up on Apple silicon. Every feature is free for 7 days, no card and no account, and after that it is bought once rather than rented.
- Shipped
- v1.0.0
- Platform
- macOS 14+
- Data
- On device
- Price
- One-time
- Updates
- Sparkle
- Built
- On a fork
copycat.deeproduct.org
Apple silicon
No account, no cloud, no sync
From £1.99, after 7 days free
Signed appcast, in-app
New interface, inherited engine
After it shipped
Every bug and request goes on a public board and is ranked by vote, so the thing most people are actually stuck on is the thing built next. It runs on the support board.
Interactive Product Experience
Welcome. Add any problem you ran into.
Open
Nothing is submitted from here. Your requests clear on refresh.

15
Reflection and next steps
What I'd carry forward
01
Don't rebuild what works
Forking Maccy's engine let all the time go into the part people see.
02
Recognising is the job
Every clipboard manager remembers. Making copies easy to recognise was the gap.
03
Ask only what it has earned
No card for the trial and no subscription after it. A small tool earns its price.
Next steps
- 01Let the panel take other shapes, like a notch tray or a floating window.
- 02Build what people vote for on the support board.
- 03Run a usability study to replace this review.
- 04Close the 2 heuristics that are only partly met.
7 days free, then buy it once.
Try it
Great products happen when the right people meet.
Let's work together
