Reelic · Case Study
Thanks for Visiting
Reach out
Overview
The short version · ~8 min read
You saved a few thousand reels. Can you find the one you need?
The problem was never saving. That is one tap and it always works. It is looking: a few thousand saves sit in a folder with no search box over it.
My solution was to index them in your own browser, over 4 key decisions that each cost something. The outcome is a folder that answers, from whatever you happen to remember, on Instagram, in a side panel or in a dashboard. The output is live at 1.0.5 in 6 browsers, on about 10 machines, with no announcement yet.
- 10+
- Beta testers so far, before any announcement
- 93+
- Every component, state and icons designed
- 2 weeks
- From idea to product
Search any saved reel.
01The outcome
It reads the caption, the hashtags and the creator together, so a reel comes back from whatever you happen to remember about it. Alongside the results it shows what the library is made of.
Interactive Product Experience
Click it
What else it does
The whole library counted on one screen: how much you saved, who from, what it was about, and what a normal week looks like.
4 key decisions
02The solution
There is no API for the thing I needed, so every one of these was a trade. 3 of the 4 made the build harder on purpose.
Where the index lives
Not taken
On a server
Shipped
On your own machine
How I got there
I started from what I would need to believe before installing this myself: that nobody else can read the folder.
What it cost
- User
- No phone version, and nothing syncs.
- Technical
- Search, storage and charts all run in a tab.
- Business
- No accounts, so no list and no retention data.
How the saves get collected
Not taken
Scroll the whole grid
Shipped
Ask for them directly
How I got there
The first build scrolled the page and read whatever loaded. It was slow, it broke every time Instagram changed, and I threw it away.
What it cost
- User
- The first sync still takes minutes.
- Technical
- A full rebuild, inside a page that blocks scripts.
- Business
- It rests on an address Instagram never promised.
Where the search should sit
Not taken
One dashboard
Shipped
3 surfaces, 1 index
How I got there
The dashboard was built first and it was the wrong place to start. Nobody leaves Instagram to search Instagram.
What it cost
- User
- 3 places to learn rather than 1.
- Technical
- 3 interfaces and 3 sets of states, solo.
- Business
- 3 surfaces to fix every time Instagram moves.
What the first sync says
Not taken
A spinner
Shipped
A live count
How I got there
This one is not a win. The first sync is slow, and showing it makes the worst moment of the product the first one you see.
What it cost
- User
- You watch a slow job instead of being distracted.
- Technical
- Search has to answer from a half built index.
- Business
- The riskiest minutes of a new install are visible.
Design System
03Design system
Reelic sits on Instagram's own dark ground and had to belong there without pretending to be part of it. Loud reads as a hijack, invisible reads as broken, so one orange does all the signalling.
Brand colours
Typography
Manrope
ABCDEFGHIJKLMNOPQRSTUVWXYZ
abcdefghijklmnopqrstuvwxyz
0123456789
Manrope carries every heading, number and lockup. Inter does body and interface. JetBrains Mono takes anything that has to line up in a column.
Icons
20 glyphs cover all 3 surfaces. Every one sits in the same 24 box, on 2 stroke weights, and inherits the colour of whatever it sits in.
Logo making process
Session 1
Iterated the name options to find the one that suits it best.
From Concept to Deployment
04How it got built
It started in a notes file with my own saved folder as the problem, and ended on the Chrome Web Store. The marks above are every tool that touched it.
It shipped. That was the point.
05The output
Live at version 1.0.5, in 6 browsers, about 2 weeks from empty folder to installable. It has also been live for weeks with about 10 users, because I built the product before I built any way to find it.
- Shipped
- v1.0.5
- Browsers
- 6
- Surfaces
- 3
- Search
- Offline
- Data
- On device
- Built
- Solo
Chrome Web Store
Chrome, Edge, Brave, Arc, Opera, Vivaldi
In page, side panel, dashboard
Fuzzy, weighted, 4 fields
No account, no upload
Design and code, start to store
After it shipped
Pain points tracked in public and ranked by vote, so the problem people feel most is built next. It runs on the support board.
Interactive Product Experience
Welcome. Add any problem you ran into.
Open
Nothing is submitted from here. Requests clear on refresh.
What gets measured from the announcement
User
- Installs that finish the first sync
- Searches per person per week
- Still searching after 7 days
Business
- Listing views that become installs
- Uninstalls inside the first 7 days
- Requests and votes on the support board
It takes one click and no account.
Try it
Great products happen when the right people meet.
Let's work together