FLOWR
Added the crossfade YouTube Music is missing. I designed it, built it with Claude and launched a paid tier. Around 140 people use it.
See it live- ROLE
- Design, build, launch
- TOOLS
- Figma, Claude, JavaScript, Chrome MV3, Stripe
- USERS
- ~140
- STATUS
- Live, paid Pro tier
- ~140
- people using it
- 14
- iterations to get the fade right
- 7 days
- free Pro trial, then one-time price
01
The problem
YouTube Music has no crossfade. Every song ends in a gap or a hard cut, which gets old fast when you listen for hours.
There were crossfade extensions already, but the most used one sits at 2.6 stars. The gap was still there to fill.
02
Designing the experience
I designed the popup like a small piece of audio hardware: a screen that shows the waveform, two big buttons for FLOW and STOP, and one blend slider set to 6 seconds by default, which sounds smooth without dragging.
Everything else (trim, lead, wash and smart outro) sits in Pro, so the free version stays a two-button thing.
03
Prototyping with AI
I built the fade engine with Claude, testing selectors and timing straight in the console on YouTube Music. It took 14 versions before the fade could start before the player skipped ahead, without any audio bleeding through.
04
Listening to users
Early users reported two bugs: the logo sometimes didn't load, and about a second of audio bled through before the fade started. I fixed both in the next release.
Around 200 people install it a month and about 140 use it every week, so keeping people matters more than finding them. An uninstall survey and anonymous usage stats now show where they drop off.
One fix came straight from that: on non-English YouTube Music, FLOWR couldn't find the Next button, so those listeners got silence between songs. Version 1.5 fixed it.
it´s free , and it works , for more or less it is simple and enough for the job , i am using a chromebook so it is a nice extension
Reback InTime, 5 stars, Chrome Web Store, June 2026
05
Going paid
Pro launched with a 7-day trial and a one-time price, plus a lower price for people whose trial ran out. Everyone who installed before launch became a founding user and kept every feature they already had.
Survey answers then said the price was too high, so I cut it to $4.99 lifetime ($3.99 win-back) and only show the buy button in the last two days of the trial.
06
What I'd do next
Use the new stats funnel to see where installs stop, and get people to their first crossfade sooner after they install.